TianZi Asia
TianZi Biodiversity Research & Development Centre


home | e-mail us | ask professional expertise

Biodiversity protection, sustainable crop production, and fair trade in China and Asia's Mekong Region
tropical rain forest flowers, exotic rare plants and herbs, products for gardening, horticulture, landscaping, health, hobby, and on wholesale

shop in our store and contribute to biodiversity protection and fair trade

go to:

Species list | Exotic plants shop | Tropical crops | Books | Puer tea | Health & aroma | Herbal medicine | Heat treatment

invest | contract grow with us | become a partner | wholesale



Feeding behavior of Sichuan takin

(Budorcas taxicolor)


by G.B.SCHALLER1, DENG Qitao2, PAN Wenshi3, QIN Zisheng4,

WANG Xiaoming4, HU Jinchu4 and SHEN Liming2


1 Wildlife Conservation International, New York Zoological Society, Bronx Park, New York 10460, U.S.A.

2 Tangjiahe Natural Reserve, Qingchuan Country, Sichuan, China

3 Zoology Department, Beijing University, Beijing, China

4 Biology Department, Nanchong Normal College, Nanchong, Sichuan, China.


Summary.-Herds of Sichuan takin in China’s Tangjiahe Reserve usually ranged in size from 10-35 with a maximum of about 100, animals generally being found on forested slopes at 1500-3000 m in elevation. Some adult males, singly and in twos, descended into valleys below 2000 m in September, remained all winter, and moved back up the slopes with the phenological stages of growth in April and May. Takin are generalist herbivores, primarily browsers. Of the 138 food species collected, most were forbs and leaves from shrubs and trees. Being large and robust, takin may push over or break saplings 8-10 cm in diameter to reach browse. Spring to autumn foods had a relatively high crude protein content(mean 10-17%), a moderately good essential amino acid balance (methionine is limiting ), and a high proportion of cell content (protein, starches, and other cell solubles) which, unlike the structural components of the cell wall(cellulose, lignin, and others),an animal can digest rapidly and almost completely. Winter food, by contrast, consisted mostly of twigs and of evergreen leaves from woody species which are low in protein(mean 7%) ,often high in secondary compounds, and otherwise of poor nutritive quality. Some animals lost condition on this diet and 2 old males died of malnutrition.

Résumé.- Les troupeaux de takins du Séchuan, dans la Réserve Tangjiahe de Chine, comprennent de 10 à 35,et jusqu à 100 individus au maximum; ils se trouvent surles versants de montagnes couverts de forêts à 1500-3000 mètres daltitude. Certains males adultes, seuls ou par deux, descendent dans les vallées au-dessous de 2000 mètres en septembre. Ils y restent tout Lhiver, et ils remontent en montagne en accompagnant le développement printanier des plantes en avril et en mai. Les takins sont des herbivores qui se nourrissent de feuilles. Parmi 138 espèces consommées, la plupart étaient de lherbe et des feuilles darbustes et darbres. Grands et robustes, les takins peuvent pousser ou casser de jeunes arbres de 8 à 10 centimètres de diamètre pour en manger les feuilles. Au printemps, en été et en automne les feuilles renferment un taux de protéine brute relativement bon (en moyenne 10-17%), une proportion dacides aminés assez bonne (la méthionine est en petite quantité) et une proportion de divers contenus de la cellule (protéine, amidon et autres substances solubles) qui, par opposition avec les constituan structuraux de la paroi des cellules (cellulose, lignine, etc.), peuvent se digérer rapidement et presque complètement. La nourriture dhiver, par contre, se compose pour la plupar de brindilles et de feuilles de plantes à feuilles persistantes despèces ligneuses, qui contiemnent peu de protéines (en moyenne 7%), souyent beaucoup de composés secondaires et généralement dune faible qualité nutritive. Certains animaux ne peuvent pas se maintenir en condition avec cette alimentation et deux males agés sont morts de malnutrition.


The takin (Budorcas taxicolor) is a robust and cow-like animal with sturdy legs, a bulging convex muzzle, and short massive horns that sweep out and back. It belongs to the subfamily Caprinae, and ,together with the muskox (Ovibos moschatus), to the tribe Ovibovini. Confined to forested mountain slopes, takin occur sporadically from Bhutan eastward through Assam and northern Burma into China. Four subspecies have been recognized of which two-the Sichuan takin (B.t.tibetana) and golden takin (B.t.bedfordi) -are wholly confined to China where they have a limited distribution(Fig .1). The two most western subspecies are dark brown; the Sichuan takin is straw-colored with splotches of gray-black on legs, back, and rump; and the golden takin is yellowish except for a black muzzle.

Until recent years , most knowledge about takin behavior in the wild was based on a few casual observations (see Schaller 1977).Two populations were observed in China during the 1970s, one of golden takin in the Qinling Mountains (Wu and Hu 1980; Anon.1981; Wu et al.1983), and one of Sichuan takin in the Qionglai Mountains (Expedition of Rare Animal Species in Sichuan 1977; Hu 1981). Between March 1984 and February 1985, we observed takin on occasion while studying giant panda (Ailuropoda melanoleuca) in the Tangjiahe Natural Reserve(32°35'N, 104°45'E) in the Min Mountains of northern Sichuan (Fig.1),and obtained new information on their food habits and feeding behavior.


Tangjiahe extends over about 400 km2 of rugged mountains at elevations of 1200-3800 m. The vegetation shows a vertical zonation. An evergreen and deciduous broadleafed forest occurs below 1700 m with such trees as Lindera, Cyclobalanopsis, Fagus, and Quercus prominent. Between 1700 and 2100 m a mixed coniferous and broadleafed deciduous forest predominates. Pinus, Tsuga, and Picea are the main conifers, and Betula, Prunus, Acer, Viburnum, and Salix are conspicuous among the broadleafed shrubs and trees; Rhododendron is common in the understory , as is bamboo of several species. A subalpine coniferous zone begins at about 2100-2300 m, depending on exposure. Tsuga and Pinus are the main conifers at low elevations and Abies above 2400 m; rhododendron and Betula are the principal broadleafed trees. At its upper limit, forest may grade into tussock grassland at only 2500 m but usually timberline is at about 3200-3300 m.

Human activity has greatly modified the vegetation in many areas below 2200 m. Most relatively gentle slopes were at one time cultivated, and former fields are now overgrown with secondary vegetation in various stages of regeneration, from meadows covered with Artemisia and other forbs to dense thickets of shrubs and saplings(Populus, Salix, Spiraea, Rubus). In 1965 a road was built into each of the 2 main valleys of the reserve, and timber was extracted in accessible areas from then until Tangjiahe became a reserve in 1978. About 300 villagers still reside in the reserve and till the slopes , all at low elevations in the eastern part.

Climate is temperate, our research base, located at 1760 m, registering an absolute minimum of -11 and a maximum of 30. January was the coldest month (mean daily minimum -5.3 and maximum 1.8), July and August the warmest (mean daily minimum 13.1-14.0 and maximum 22.6-24.7).The first spring flower was seen on 18 March, the last snow fell a few days later, and throughout April the mountains greened rapidly. Between May and October there was a rainy season when 93.5% of the total 12-month precipitation of 1130 mm fell . By mid-September the first yellow leaves of autumn were seen and most deciduous trees were bare a month later. December 14 brought the first heavy snow to the valleys.


We observed takin mainly in the Beilu valley along 12 km of road at elevations of 1500-2000 m . Solitary takin males often foraged low on slopes where we could easily watch them; several animals were habituated to people, permitting approach to within about 10-15 m .Herds were infrequently observed. Selected takin foods were oven-dried and later analyzed for crude protein(calculated as total Kjeldahl nitrogen_6.25)and essential amino acid content (using high-pressure liquid chromotography with a Beckman 121 MB Amino Acid Analyzer). To obtain an indication of nutrients available to takin, the relative proportions of constituents in cell wall and cell contents of 7 foods species were chemically analyzed by the Van Soest (1982) method.


A number of adult male takin descended into the Beilu valley in late September where we often saw them until April. In late April , they began to move up the slopes, and after May usually remained in the forest above 2200 m until autumn; no takin were seen near the valley floor between May 22 and September 17. The retreat of males coincided with the upward advance of new plant growth in spring. Animals were up high during the mating season which is said to extend from late July to early September(Wallace 1913).Of 288 male takin tallied alone or in male herds, 87.5% were solitary, 8.3% in twos, and 4.2% in threes; West(1926) reported one group of 4 male takin in India .Associations between males appeared casual in that animals seldom remained together for long .Solitary males often foraged for several days within an area of 1 to 2 ha. One bull was observed on 60 days during the year, always alone except on one occasion. In spring .he foraged mostly in a former field, whereas in autumn and winter he roamed along about 6 km of valley (Fig.2).

Females and subadults, as well as some males, associated in herds, which, unlike solitary males, traveled much and seldom descended into the Beilu valley. With animals scattered in the forest, we found it difficult to determine herd size, but most contained 10-35 individuals ,figures similar to those given by Wu and Hu(1980) and the Expedition of Rare Animal Species in Sichuan(1977);Yin(1967) reported a herd with up to 300 animals in Burma. In May, we observed one large herd of 93-112 takin. Twenty percent of these animals were small young, dark brown in color, born the previous month. This herd apparently split up latter in the year; at least in late October and early November, 3 to 4 herds totaling about 100 takin were observed in the same area as they descended in late afternoon to the Beilu River where some animals drank.

The size and composition of 3 separate herds in November showed that adult females outnumbered adult males by about 2:1; other males were solitary or in male herds. Young comprised 20% , the same figure as in May, and yearlings also 20% ,indicating low mortality in these age groups(Table 1).We found no evidence of local people hunting takin, and the 2 potential predator species –leopard (Panthera pardus) and Asiatic wild dog (Cuon alpinus) -were scarce. The content of 120 dropping of these 2 predators (we could not distinguish the droppings in many instances), expressed in frequency of occurrence, consisted of 72.5% takin , 18.3% tufted deer (Elaphodus cephalophus), 8.3% muskdeer (Moschus berezovski), and 0.8% golden monkey (Rhinopithecus roxellanae). Although takin was the major food item, some of the meat had been scavenged from animals dead of malnutrition and other causes.


Table 1. Composition of 3 takin herds in the Tangjiahe Reserve



Subadult male

(2.5 yrs.)

Adult and

Subadult female


(1.5 yrs.)


(0.5 yrs.)






















Takin forage on a wide variety of plants. Hu(1981) listed 40 food species for the Wolong Reserve in the Qionglai Mountains, and Wu et al.(1983) tallied 63 in the Qinling Mountains. We noted 138 food species in Tangjiahe Table 2), and detailed work would no doubt raise this figure. Takin are primarily browsers. Our list of food species includes 7 vines, 52 herbs, and 70 shrubs and trees but only 8 grasses and grass-like plants. With so many plants palatable, takin often have a wide selection available at a site: one herd foraged on 27 species while crossing an old field.



TABLE 2. Plant species eaten by takin in the Tangjiahe Reserve.

Grasses and grass-like plants

Phlomis sp.(S-6.6)


Polygonatum cirrhifolium

Carex sp.

Polygonum lapathifolium

Deyeuxia scabrescens


Fargesia denudata*(S-15.8)



Prenanthes henryi

Millium effusum


Roegueria kamojii

Primula moupinensis(S-4.4)


Rheum palmatum


Rubia cordifolia(S-14.3)

Equisetum hiemale*(S-11.4)

Rumex japonicus


Sambucus adnata


Sanicula chinensis(S-30.0)

Lunathyrium garaldii

Sarssurea sp.


Sedum sp.


Smilax stans

Actinidia profunderum

Stellaria infracta

Celastrus sp.

Thlaspi arvense

Lonicera henryi*(W-8.4)

Shrubs and trees


Vicia cracca(S-12.2)

Rubus ichangensis*


Sabia latifolia

Acer caesium

Vitis sp.




Agrimonia pilosa


Androsace henryi


Anemone hupehensis

Actinidia kolomikta(S-24.3)

Angelica sp.(A-6.7)

Aralia chinensis(S-17.8)

Artemisia moorcroftiana(S-16.7)

Betula luminifera


Buddleia davidii(W-15.6;Su-13.8;A-8.2)

Aster ageratoides

Choerospondias axillaris

Astilbe chinensis

Corylus sp.(A-17.6)

Boehmeria clidemoiodes

Cotoneaster ambiguus

Brassica campestris(S-4.4)


Cacalia profunderum


C.tanqutica (S-17.9;A-10.2)

Cyclobalanopsis oxydon*(W-7.2)

Caltha fistulosa

Dacaisnea fargesii(S-18.3)

Cardamine tangutorum(S-8.1)

Daphne tangutica

Cardiocrinum giganteum

Deutzia glomeruliflora (S-22.7)

Cisium leo (A-14.9)

Dipteronia sinensis

Corydalis edulis (S-19.9)

Elaeagnus lanceolata*

Cucubalus baccifer

Enkianthus chinensis

Glechoma biondiana

Euonymus alatus



Heracleum scabridum(S-17.0)



Ficus heteromorpha

Heterolamium sp.

Helwingia chinensis*

Impatiens diceatra



Hydrangea sp.(S-22.2)

Lamium barbatum

Hypericum patulum

Laportea macrostachys(S-18.9)

Hex pernyi*

Ligularia intermedia(A-11.4)

Juglans cathayensis


Kerria japonica

Paris thibetica(S-13.9)

Lindera communis*(W-8.8)

Petasites tricholobus(S-19.3)

Litsea cubeba



Peucedanum sp.


Lonicera pileata*(W-7.5)


Maddenia hypoleuca


Meliosma flexuosa

Ribes glaciale

Morus australis

Rosa omeiensis

Neillia sinensis

Rubus coreanus(S-13.7)

Photinia lasiogyna*(W-9.3)


Pinus armandii*

Sambucus adnata(A-17.5)

Populus davidiana

Salix eriostachya



Prunus brachypoda





Spiraea japonica(S-14.8;A-10.8)

Quercus glandulifera

Syringa sp.


Ulmus sp.

Rhododendron argyrophyllum*

Viburnum betulifolium




The plants were identified by Qin Zisheng. An asterisk (*) indicates an evergreen species. The information in parentheses represents the percent crude protein in leaves and the season in which the Sample was collected. Winter(W) samples were collected from November 19 to March 20,Spring (S) ones from April 17 to May 18,summer(Su) ones from July 30 to August 17, and autumn(A) ones from September 12-22.

Food habits changed seasonally. During winter from November to March, takin browsed much on evergreen broadleafed species, favoring a vine (Lonicera) and certain trees (Rhododendron, Lindera, Photinia) and shrubs(Eleagnus, Lonicera, Buddleia). Of these Buddleia , an abundant shrub along the valley floor, was important to takin males wintering there. At elevations below 2000 m , Buddleia shed old leaves in autumn but retained young leaves through the winter, providing takin with browse .The leaves of some evergreen broadleafed species were eaten but not favored, among them Cyclobalanopsis, llex, and bamboo. Branch tips from deciduous trees and shrubs (Prunus, Daphne, Juglans, Dipteronia, Viburnum and others) were also taken; Salix bark was sometimes gnawed or stripped off ,as was Pinus bark. Hu(1981) and Wu et al.(1983) noted that Abies bark may also be consumed. Most herbs were dead in winter or covered with snow; we saw no evidence that takin paw away snow to uncover food. One tall herb, Artemisia subdigitata, was a major food plant of males in March, animals eating both dried leaves from stalks and green ones that overwinter at the base of the plant.

During the first half of April many herbs appeared below an elevation of 2200 m. There were luxuriant herb meadows on shaded stream terraces, on some gentle forested slopes ,and on recently abandoned fields. Takin concentrated there to eat the sprouts of Aster, Polygonum, Sedum, Paris, Cardamine, Ligularia, Glechoma and others with a particular liking for Rubia, Brassica , and Cacalia. Grass was not much sought at this time except for Roegueria, A species of forest clearings. As soon as available, leaves of the much-favored Salix , Aralia, Deutzia , Litsea, Ficus, and Morus were browsed, as were those of many other shrubs and trees. Evergreen broadleafed species, so important as winter feed, were largely ignored from sping through autumn.

Herbs and leaves provided takin with the bulk of their diet from June through October as well. In September, when males returned to the lower slopes of the Beilu valley, they favored Polygonum, Ligularia, Cacalia, Angelica and other herbs, as well as the leaves of Spiraea, Vitis , Sambucus, and Cotoneaster. Salix and a few other deciduous shrubs retained some leaves into winter, and these takin sought when all but evergreen leaves became scarce.

Takin ate soil at certain sites, behavior especially prevalent from late September to early November. One small herd repeatedly licked and gnawed the mud bricks of an abandoned hut until one wall became so thin that it collapsed. Soil may provide required minerals or somehow help detoxify plants and otherwise assist digestion (Kreulen 1985). Cooper (1923) and Ward(1921) also observed takin at soil licks.



Broad, flexible lips enable takin to forage efficiently and selectively. An animal may pluck the terminal leaves off a twig, or merely slide a branch through its mouth ,collecting all leaves in a wad. Its large body size gives takin access to food resources beyond the reach of other mountain ungulates in Tangjiahe. One male we measured had a shoulder height of 126.5 cm and another of 140 cm, indicating that animals can with ease reach browse at a height of 2 m .A takin may bite into a branch and snap it with a twist of the head, thereby gaining easy access to the leaves. Browse beyond reach was obtained in one of several ways. An animal may bend a sapling or shrub with side of neck or head, and, holding it down with chin or foreleg, feed on the branches. Or it may rear on its hindlegs, often with one or both forelegs propped against the trunk ,and browse as high as 2.5 m above ground. Using its weight ,a takin sometimes leaned into a tree or shrub and bent or broke it .If the stem did not snap, the animal prevented it from springing erect by straddling it or holding it down with foreleg or neck while browsing. Trees up to 8-10 cm in diameter may be broken in this manner.

Whenever we encountered solitary males or male herds in daytime, we noted whether the animals were feeding. Males foraged primarily in the morning and late afternoon, tending to lie down and rest at midday (Fig.3). Activity changed somewhat with the seasons. In March and April, takin seldom foraged between 0900 and 1500 h, whereas from November to February, when food was most scarce, they were sporadically active at midday as well. When there was snow cover, animals usually rested while standing up, their body often broadside to the sun.



As many herbivore studies have shown, food preferences are related to the nutritive quality of forage (Longhurst et al.1968; Belovsky 1978), animals often selecting for those plant parts that have the most protein and digestible energy, and least amount of cellulose. We chemically analyzed spring, autumn, and winter forage to document differences in nutritive quality during seasons of abundance and scarcity. Water content in forage is also of importance to herbivores. If content is high, animals must eat much bulk to obtain few nutrients-and the amount an animal can eat is limited by the capacity of its digestive tract (Westoby 1974).

Spring feed was succulent, containing 80-90% water. Herbs retained a high water content through autumn, but levels in leaves of shrubs and trees decreased. Winter forage, with its high proportion of twigs and leathery evergreen leaves, had only a moderately high content averaging 55%(Table 3).


TABLE 3 .Water and crude protein content of takin food plants by season(mean±standard deviation).



(30 species)


(12 species)


(12 species)

% Water


Shrubs and trees

All foods combined










% Crude protein


Shrubs and trees

All foods combined











Crude protein in spring feed varied fron as low as 4% in 2 herbs (Primula, Brassica) to as high as 22-30% in 4 species (Actinidea, Deutzia, Hydrangea, Sanicula) with a mean of 16% for 18 herbs and 12 shrubs and trees analyzed. By autumn, mean protein levels had dropped, though not significantly, to 11% is 5 herbs and 7 woody species tested. These levels ranged from 6% in Cotoneaster, Prunus and Angelica to 17% in Sambucus, Corylus, and Salix. With a mean protein content of 10%,12 winter food plants had values similar to autumn ones(Table 3). However, this figure is somewhat misleading. Three species of bamboo leaves, not favored as food, averaged 15% protein, whereas 7 other broadleafed evergreens which contributed much to the diet averaged only 7%.

Eight amino acids are considered to be essential for optimum growth; we analyzed takin foods for all but tryptophan (Table 4). Methionine was present in the smallest amount in all samples with the exception of 2 spring foods (Deutzia, Salix)in which threonine took that position. This suggests that methionine represents a limiting amino acid in the diet. The nutritional value of a plant lies, however, more in a good balance of essential amino acids than in the amount of any one .Egg is often used as a standard because of its good balance. The ratio of the least adequately supplied essential amino acid in a plant and a standard reference protein with the best balance (egg) represents the amino acid score (see Schaller et al.1985). If egg is given a score of 100, then relative food values can be determined (based on a methionine value of .342 and threonine of .310 in egg). In spring, Petasites(83), Sanicula(94), Salix(158) and Deutzia(316) had the highest scores, and Acer(8), Prunus(13), Brassica(13), and Primula(15) the lowest; most others scored between 30-60. Autumn scores were less variable than spring ones ,most samples ranging from 40-70,with Cotoneaster(35) the lowest and Corylus(98) the highest. In winter, scores were low, the amino acid balance poor, with Salix bark(12) having the worst score and green Buddleia and Artemisia leaves(43) the best.

The cellulose and other structural components in the cell wall of browses tend to be relatively indigestible if lignin values are high (Robbins and Moen 1975).We compared the components of cell wall and cell contents (protein, starches, lipids) of 6 favored herbs eaten in spring with those of Fargesia scabrida bamboo which takin eat only sparingly in winter (Table 5). Proportions of cellulose and lignin were similar in herbs and bamboo. Among the cell solubles, mean crude protein content in herbs was double and the remaining cell contents triple that in bamboo.



To maintain an adequate nutritional level, a herbivore must forage on those species which at the time offer the most protein and digestible energy for the least amount of effort (Westoby 1974; Belovsky 1978). Young growing plants generally have a higher protein content than adult and dormant ones, with forage containing less than 5% crude protein being too low in quality to maintain body weight(Sinclair 1977).A herbivore improves its chances of obtaining a high-quality diet if it finds many species palatable and if it shifts to different feeding areas as necessary. However, among the many plant species in a habitat only a limited number are both available and preferred as forage by an animal. Palatability is not always correlated with nutrient content, some species of high quality being avoided. Such avoidance may be caused by the presence of secondary compounds which can either be toxic or inhibit the growth of rumen microorganisms and thereby impede digestion(Schwartz and Hobbs 1985).An animal can tolerate high levels of secondary compounds only in moderate amounts, for its ability to degrade toxins in a given time is limited (Longhurst et al.1968).Another factor affecting forage preference is lignin content, because a high lignin level in the cell wall may decrease cellulose digestibility (Robbins and Moen 1975). As Freeland and Janzen(1974) noted, herbivores "should prefer to feed on foods that contain small amounts of secondary compounds, and their body size and searching strategies should be adapted to optimize the number of types of foods available ." How well do takin follow these predictions?

With their large size, seasonal shifts up and down slopes, and dietary flexibility, takin have a remarkably wide choice of food plants from which to select an optimum diet; they are true generalist herbivores. In spring, new growth provides an ample, nutritious, and mixed diet seemingly low in secondary compounds. At that season,50% or more of the nutritive content of plants consists of protein, starches and other cell solubles which an animal can digest rapidly and almost completely(Van Soest 1982). In addition, such high-quality forage enhances the ability of microorganisms to ferment cellulose (Belovsky 1981). An animal, therefore, needs to allocate little time to digest such feed. Since spring forage is succulent, providing much bulk but relatively few nutrients, and rumen capacity is limited, rapid passage is essential. The amino acid balance in browse is variable. Some with a poor balance (Primula, Brassica) were for unknown reasons favored as much as some with a good balance (Deutzia, Salix). However, preference for one species may be affected by availability of other species. For example, in early spring Primula was locally the most abundant forb in the forest. And ,as Belovsky(1978) has shown for moose (Alces alces), animals may consume certain plants for specific minerals rather than for protein and other nutrients.

In summer and autumn, takin retained their varied diet, and many species showed protein values and amino acid scores as high as those in spring. Conditions , however, changed in winter. Deprived of most herbs and deciduous leaves as a main source of nutrients, takin had to switch to less palatable plant parts either of the same or different species. Protein in Salix leaves was, for example, high(12-18%) from spring to autumn, but only twigs (8%) and bark (5%) remained as primary winter feed. Twigs from deciduous shrubs and trees browsed by white-tailed deer(Odocoileus virginianus) in winter in New York had a cell wall content of 45-65% of which 15% was lignin(Robbins and Moen 1975); it is likely that levels in Tangjiahe were similar. With the cell wall of such browses relatively low in digestibility because of a high lignin content, takin should prefer leaves. This they did .At low elevations, males sought the few nutritious (15.6% protein) Buddleia leaves that over-winter. However, most available leaves were from evergreen shrubs and trees with a mean protein level of 7%, barely high enough to sustain takin. Furthermore, several browse species –Rhododendron, Lindera, Photinia are high in secondary compounds, judging by their aromatic odor, which may prevent animals from fully assimilating proteins (Yalden 1978).Takin may have a fair tolerance for certain inhibitory species such as Lindera, Photinia and Artemisia, the leaves of which are much eaten, but they may have a low tolerance for the presumed phenols in Quercus and Cyclobalanopsis and essential oils in Pinus, browses of low palatability. In sum, forage was abundant but of low quality, making winter a lean season for takin. Males in the valley visibly lost condition on this diet, and 2 old animals, their molars heavily worn and marrow fat exhausted, apparently died of malnutrition, one weighing 208 kg and the other 259 kg.

Given a takin’s tolerance for varied browse and the shortage of nutritious winter forage, it puzzled us why animals ate little bamboo. The leaves are evergreen, abundant, available, and nutritious ,averaging 15% protein in 3 species analyzed; the lignin: cellulose ratio is no less favorable than in other browse;and secondary compounds are unlikely to be present in toxic amounts. Silica (SiO2),a structural component of cell wall often complementing lignin, is known to inhibit digestion at high levels (Van Soest 1982).Most browse contained little silica, averaging 0.2%. in contrast to F.scabrida bamboo with a monthly average of 7.4% in leaves (Schaller et al. in press).But this level may not be high enough to affect palatability.

Past agriculture and timber extraction have resulted in much secondary growth in various stages of regeneration. This habitat will provide takin with forage in great abundance and variety until forest succession reduces production. The high proportion of subadults in the population reflects this good nutritional base. No census of takin has been made, but we estimate that 350-450 frequented the reserve in 1984. Protected now from human predation, this population can be expected to grow until the amount of winter browse becomes a limiting factor.



Wang Menghu of the Ministry of Forestry, Beijing; Hu Tieqin, Bi Fengzhou, Gung Tongyang, and Fu Chengjun of the Forestry Bureau, Chengdu; and Yue Zhishun and Jiang Mingdao of the Tangjiahe Natural Reserve helped make our work possible.K.Schaller, Wang Fulin, and K.Johnson assisted in the field. Protein and amino acid analyses were done at the Chemistry Department, Beijing University and other determinations at the Department of Animals Sciences, Cornell University, through the courtesy of J.Robertson. Research at Tangjiahe represented a collaborative effort between China’s Ministry of Forestry, World Wildlife Fund, and Wildlife Conservation International, a division of the New York Zoological Society. We are grateful to these individuals and institutions for their support.



ANONYMOUS.1981.- Natural investigations of and recommendations about the Tsashui takin natural reserve, Shaanxi Province. Shaanxi Province Scientific and Ecological Committee, Xian.(In Chinese).

BELOVSKY.G.,1978.- Diet optimization in a generalist herbivore, the moose, J.Theoretical Population Biol.,14:105-134.

BELOVSKY.G.,1981.- Food plant selection by a generalist herbivore: the moose.Ecology,62:1020-1030.

COOPER.H.,1923.- The Mishmi takin (Budorcas taxicolor). J.Bombay Nat.Hist.Soc., 29:550-551.

Expedition of rare animals species in Sichuan, 1977.- A report on the rare animal species in Sichuan. Sichuan Forest Bureau, Chengdu. (In Chinese).

FREELAND.W., and D.JANZEN.1974.- Strategies in herbivory by mammals :the role of secondary compounds.Am.Nat.,108:269-286.

Hu Jinchu, 1981.- Ecological and biological studies of panda, golden monkey and takin in Wolong Natural Reserve. Sichuan People’s Publ. House, Chengdu.(In Chinese).

KREULEN.D., 1985.- Lick use by large herbivores: a review of benefits and of soil consumption. Mammals Rev.,15:107-123.

LONGHURST.W., H.OH.M.JONES. and R.KEPNER.1968.-A basis for the palatability of deer forage plants.Trans.N.Am.Wildl.Nat.Res.Conf.,33:181-192.

ROBBINS.C., and A.MOEN.1975.- Composition and digestibility of several deciduous browses in the northeast.J.Wildl.Manage,39:337-341.

SCHALLER.G., 1977.- Mountain monarchs: wild sheep and goats of the Himalaya. Univ. Chicago Press, Chicago.

SCHALLER.G., Hu Jingchu, Pan Wenshi and Zhu Jing,1985.-The giant pandas of Wolong .Univ.Chicago Press, Chicago.

SCHALLER.G., TENG Qitao, K.JOHNSON.Wang Xiaoming , Shen Liming, and Hu Jinchu, in press.- Feeding ecology of giant panda and Asiatic black bear in the Tangjiahe Reserve, China. In:Carnivore behavior, ecology, and evolution, J.GITTLEMAN ed. Cornell Univ.Press, Ithaca.

SINCIAIR.A., 1977.-The African Buffalo.Univ.Chicago Press,Chicago.

SCHWARTZ.C., and N.HOBBS.1985.- Forage and range evaluation.Pp.25-51 in : Bioenergetics of wild herbivores, R.HUDSON and R.WHITE.eds.CRC Press, Boca Ratan.

VAN SOEST.P., 1982.- Nutritional ecology of the ruminant. O and B Books, Corvallis.

WALLACE.H., 1913.- The big game of centraL and western China .John Murray, London.

WARD.F., 1921.- Some observations on the birds and mammals of Imaw Bum.J.Bombay Nat.Hist.Soc., 27:754-758.

WEST.E., 1926.- Takin shooting in the spring. J.Bombay Nat.Hist.Soc., 31:273-275.

WESTOBY.M., 1974.- An analysis of diet selection by large generalist herbivores.Am.Nat.,108:290-304.

WU Jiayian, HAN Xiping and Den Fengming,1983.- The takin and its protection.Wildlife,1:14-18.(In Chinese).

Wu Jiayian and Hu Zhichi, 1980.- Resources investigation and geographical distribution of precious animals in Shaanxi Province. Shaanxi Province Agricultural and Natural Resources Committee Office, Xian. (In Chinese).

YALDEN.D., 1978.- Observations on food quality in Scottish red deer (Cervus elaphus) as determined by chemical analysis of the rumen contents.J.Zool.Lond., 185:253-277.

Yin.U., 1967.- Wild animals of Burma. Rangoon Gazette, Rangoon.



Table 4. Essential amino acid content(in mg/100 g) of takin food plants by season(mean

±standard deviation).1) Tryptophan not tested.Usually only one sample of each species was


Amino acid1)


























Leaf and stem









Deciduous woody plants





















Leaf and stem









Deciduous woody plants




















Artemisia aubdigitata

Dead leaf









Buddlela dayidll

Green leaf









Equisetum hiemale

Green stalk









Salix hoterochroma


Leaf and twlq

















Evergreen plants












TABLE 5. Nutrient content(in percent dry matter)of selected takin foods. Cacalia was collected

in June, Fargesia monthly in November, December, January and March,and the others in

May. The shrubs and forbs were also eaten by Asiatic black bear (Ursus thibetanus).


Plant part



Cell solubles

Cell wall

Crude protein

Remaining cell solubles




Hydrangea sp.

Aeanthopanax henry1

Petasites trieholobus

Phlomis sp.

Jleraeleum peabridum

Cacalin tangutica

Fargesin acabrida



















































spirituality : design : water : fashion
amphibia : reptilia : butterflies : panda : takin
ecotourism : Xishuangbanna : Shangrila : Yunnan : Mekong
books : Buddhist flowers : consulting : science of ecology : medicine
bananas : ginger : herbs : orchids : trees : succulents : lianas : sacred trees
Gabrielle's traditional motive Chinese jewelry : Wu Jialin's black and white photography
rain forests : river banks of the Mekong : pools : sacred forests : sinter terraces : chili growing
AIDS/HIV : asthma : acne : herpes : hepatitis : cancer : diabetes : health tea : malaria : Puer tea : orchid tea

is the Thai internet portal for biodiversity products designed to protect endangered species in cooperation with local ethnic communities
a project of the
Xishuangbanna, Yunnan, China

© 2000 - today