Skip to main content

Steps towards sustainable harvest of Ophiocordyceps sinensis in Bhutan


The insect-pathogenic fungus Ophiocordyceps sinensis (better known as Cordyceps sinensis) is harvested over much of the Himalayan plateau as a highly prized remedy in traditional Oriental medicine. Over the past 10 years its financial value has increased dramatically, with collectors paid as much as US $12,500 kg−1 for top-quality material. This is causing significant distortion to local economies, and there is widespread concern that the current rate of collection is unsustainable. This paper introduces the fungus and its insect hosts, documents some of the biological and social constraints to achieving sustainability, describes the socioeconomic climate within which harvest and sale occurs in Bhutan, and details the measures put in place by the Royal Government of Bhutan to promote wise management of this valuable natural resource.


Ophiocordyceps sinensis (better known as Cordyceps sinensis or by its Tibetan name yartsa gunbu) is one of the most highly prized and expensive natural products used in traditional Oriental medicine. It is used as a remedy for a wide range of conditions, especially to aid in building strength and recovery from illness. Several factors over the past 10 years and more have combined to increase demand, including widespread publicity surrounding its use by record-breaking Chinese athletes, health concerns caused by the SARS and bird flu outbreaks in S and SE Asia, and increasing affluence in China, its principal market. In Bhutan O. sinensis was placed on Schedule 1 of the Forest and Nature Conservation Act and its harvest was illegal until 2004 (apart from a small pilot scheme in Lunana from 2002), when a limited collection regime was introduced. Sustainability of its harvest cannot be assured without a full understanding of the fungus, its insect hosts and their inter-relationships, and a monitoring programme to establish baseline population densities. This paper describes the first stages in these processes.

The fungus

Ophiocordyceps sinensis (Fig. 1) is a fungal parasite of larvae (caterpillars) belonging to the ghost moth genus Thitarodes (Hepialidae, Lepidoptera). It has a wide but patchy distribution in montane grasslands of the Tibetan plateau, being recorded from Bhutan (Namgyel and Tshitila 2003), China (Tibet (Xizang Autonomous Region), Gansu, Qinghai, Sichuan and Yunnan provinces; e.g. Pegler et al. 1994; Winkler 2005), India (Arunachal Pradesh, Himachal Pradesh, Sikkim, Uttarakhand; Sharma 2004; Winkler 2005) and Nepal (Devkota 2006). Records from other parts of China (e.g. Tian Shan; Zang and Kinjo 1998) require confirmation. Its altitudinal distribution in Bhutan ranges between about 4,200 and 5,200 m, and most published records of altitude for this species in other countries fall within this range.

Fig. 1
figure 1

Ophiocordyceps species. a Living stroma of O. sinensis from Namna, north-western Bhutan; note roughening caused by developing fruit-bodies on stroma surface; b stroma of O. sinensis dug out of the ground at Namna, showing stroma developing from larval cadaver; c stromata of the forest ecotype (O. sinensis?) from the Bumthang Valley, north central Bhutan. Note remains of the cocoon surrounding the larval cadavers, which is cleaned off before sale

The species was first described by Berkeley (1843) as Sphaeria sinensis, from material sent to the UK from markets in Guangdong (Pegler et al. 1994). For many years, it has been referred to as Cordyceps sinensis following its transfer by Saccardo (1878), but recent molecular phylogenetic studies (Sung et al. 2007) have shown that Cordyceps as recognized at that time was polyphyletic, and that C. sinensis is not congeneric with the type species C. militaris. Transfer of the species to Ophiocordyceps, within the newly recognized family Ophiocordycipitaceae was therefore recommended.

There have been some studies that conclude that the Himalayan Cordyceps associated with ghost moths comprise a species aggregate rather than a single species. Zang and Kinjo (1998) list seven species of Cordyceps parasitic on hepialid larvae in the Himalayan region. The diagnostic characteristics cited were based on morphological examination, and it is probable that some at least of these taxa are based on minor morphological traits of limited phylogenetic significance. Following studies of RAPDs by Chen et al. (1999), preliminary molecular studies based on ITS analysis (Stensrud et al. 2007) indicated that several distinct clades may exist within O. sinensis, and the authors speculated that these might represent cryptic species. Our own studies on Ophiocordyceps associated with ghost moth larvae in Bhutan have not yet identified clearly distinct taxa based on morphological analysis, with the exception of a coniferous forest-inhabiting ecotype with unusually large stromata that may be conspecific with collections from China from similar habitats. These have been assigned (almost certainly incorrectly) to the European/Mediterranean species C. gracilis by Zang and Kinjo (1998). A molecular analysis of representative collections from each region of northern Bhutan is under way.

Ophiocordyceps sinensis has an anamorph (asexual stage) referable to Hirsutella, with ascospores germinating directly to produce conidiogenous cells and conidia that probably act as the infective propagules; this has been demonstrated in related species (Hywel-Jones unpublished observation). Despite claims to the contrary (e.g. Pegler et al. 1994; Li et al. 1999) the ascospores of Ophiocordyceps species are aseptate and do not fragment following dispersal in common with true Cordyceps species (Sung et al. 2007). Anamorph-teleomorph connections for this species are highly confused, with no fewer than 22 anamorph names linked to the sexual stage (Jiang and Yao 2002). Many of these are erroneous, and some are likely to have led to the use of misidentified strains in commercial preparations. Misidentifications can also be demonstrated in public sequence databases (Stensrud et al. 2007), some of which are likely to be the result of sequencing rapidly growing contaminating fungi rather than the genuine species. The genuine anamorph of O. sinensis grows very slowly in standard culture conditions, taking 1–2 months to grow to 2–3 mm diameter on standard media (e.g. Potato Dextrose Agar) at 15–20°C (Hywel-Jones unpublished observation).

The host moth

The host range and specificity of Ophiocordyceps sinensis remains uncertain. Early records indicate that the fungus occurs on Hepialus armoricanus. That species was originally described by Oberthür (1909) from a specimen reportedly caught in his garden in Brittany (hence the choice of epithet). There have been claims that the moth escaped into the garden from a hatching pupa sent to him by missionaries in Tatsienlu, Tibet (Zimmer 2002, 2006) and the story curiously features in the novelist and keen entomologist Nabokov’s work The Gift (Nabokov 1991). It seems more probable that the adult specimen was sent from Tibet and became inadvertently mixed with Oberthür’s own collections.

Hepialus armoricanus was later separated from that genus and placed within Thitarodes (Hepialidae; Viette 1968). Species of Hepialidae are commonly referred to as ghost moths, and they are thought to be amongst the most primitive families of Lepidoptera. Thitarodes species are found all over the Himalayan plateau and other temperate regions of eastern China, with a few extending as far as Myanmar, Taiwan, Japan and north-eastern Russia. Around 50 species have been described (Nielsen et al. 2000; Zhu et al. 2004), many of them in recent years, and are still referred to as Hepialus species in much of the Chinese literature. The true diversity of the genus is not easy to establish, as in a number of cases existing descriptions and illustrations are inadequate for unequivocal identification. Our studies in Bhutan indicate that at least two species of Thitarodes occur sympatrically in the montane pastures of our main research site, and a further species may well be host to the forest ecotype of Ophiocordyceps sinensis referred to above. We have not so far been able to assign any of these to previously described species, and we are confident that one at least represents an undescribed taxon (Fig. 2).

Fig. 2
figure 2

Thitarodes species. a Male adult moth caught at Namna; b female adult moth from Namna; c late-instar larva, probably of the same species; d Pupa; e Light-trapping for adult moths at Namna

Thitarodes species spend by far the greater proportion of their lives as larvae, living in the surface layers of grassland soil and feeding on plant roots. Most are considered to be non-specialists, and related groups of hepialid larvae cause economic damage to pastures. They almost certainly develop over several years (up to five according to Winkler 2005), remaining dormant throughout the winter months, pupating and emerging as adults in early summer. The adult moths only live for a short period (as little as 2–5 days in the case of ‘Hepialus’ biruensis; Chen et al. 2002). They lay large numbers of eggs by spraying indiscriminately during flight, and mate following crepuscular ‘lekking’ displays in twilight periods (Andersson et al. 1998).

Fungus–insect interactions

Many of the features of host-parasite interactions between Ophiocordyceps sinensis and Thitarodes species remain inadequately studied. We have made some contributions to this study, but the difficulties of observing the infective process in the field are considerable. In Ophiocordyceps sites in northwestern Bhutan, Thitarodes moths fly for short periods at dusk during early to late July, and we have observed oviposition of captured females. We have also studied larvae at various stages of development, following excavation of their habitat. The observed lack of clear size classes tends to support claims that they develop over several years, although the likely presence of multiple species may obscure any instar-related pattern. We have also observed pupae developing in early June.

Infective propagules of Ophiocordyceps species are short-lived and lack survival traits such as melanization and substantial nutritional reserves. They also do not form substantial mycelial networks in soil, from which further infective propagules may develop. It is therefore most probable that larvae are infected shortly after release of spores from the fungus stroma, on the surface of soil or lower layers of vegetation. In other species of entomogenous Hypocreales the fungus has been shown to infect initial larval stages by penetration of the still-soft exoskeleton, and remain dormant within the insect for long periods. This scenario is the more probable for O. sinensis as while we have found many Thitarodes larvae in the soil of our study sites, we have never observed late-instar larvae on the surface at any time of the year and thus vulnerable to infection. Infection most likely occurs at first instar stage, as their hatching period overlaps with the discharge and dispersal of the ascospores. Following a period of dormancy within the caterpillar, a yeast-like stage develops which spreads throughout the haemocoel and concentrates within the caterpillar’s lipid reserves. Growth of the fungus thus depletes the host internal nutrient supply, and the caterpillar may well be effectively killed by starvation as its energy resources become inadequate for successful pupation. Following death of the caterpillar, the fungus changes to a mycelial growth phase and absorb almost all of the remaining insect tissues within the exoskeleton (Xing and Guo 2008). There are reports in the grey literature that the fungus stroma can emerge prior to death of the caterpillar, giving the host organism a horned appearance (Winkler 2005), but we believe these to be apocryphal.

In north-western Bhutan, stromata develop from the head end of the buried caterpillar cadaver and grow through the soil layers to emerge from around the middle of May. Ascomata (containing the ascospores) may develop on the exposed stroma surface quite soon after emergence, but they do not mature fully for some 4–6 weeks, with ascospores being ejected in July and the first part of August, coinciding with the flight period and egg-laying of the host moths. It seems likely that stromata may retain the capacity to release spores for some considerable time, and release may be triggered by similar environmental conditions (e.g. temperature and/or humidity) to those that stimulate emergence of the host insects.


Pharmaceuticals derived from Cordyceps species and their relatives are used widely in western medicine, for example cyclosporin which is used to prevent rejection after transplant surgery and to treat autoimmune disorders (Borel 2002), and ergotamine which can be used to treat migraine and also for induction of childbirth (Schardl et al. 2006). Ophiocordyceps sinensis is not currently used in mainstream Western healthcare, but is one of the most highly regarded remedies in traditional Oriental medicine and has been widely adopted by devotees elsewhere. O. sinensis has been shown to produce a wide range of bioactive molecules which have been claimed to produce therapeutic effects (reviewed by Paterson 2008), although in many cases the research is not peer-reviewed to international standards. It is also unclear whether bioactive compounds are produced in sufficient quantity in the natural product to have an adequate therapeutic effect in the amounts typically consumed.

Traditional Oriental medicine is practised in a substantially different manner to western healthcare, often adopting a holistic approach where medicines are used to support and stimulate overall health rather than curing specific conditions. It is also common for a range of medicines to be prescribed simultaneously. For these reasons it is unsurprising that little direct evidence of therapeutic effect of Ophiocordyceps sinensis exists within traditional Oriental health systems, and it would indeed be difficult to devise specific trials except in very large-scale research projects.

In China, Ophiocordyceps sinensis and its larval cadaver are collectively referred to as 冬 虫 夏 草 (‘Dong Chong Xia Cao’; Jiang and Yao 2002), a translation of the Tibetan term ‘yartsa gunbu’. It is known throughout the Himalayan region by various modifications of the Tibetan name, including ‘yartsa guenbub’Footnote 1 in Bhutan and ‘yarsagumba’ in Nepal (Devkota 2006). The name translates as ‘winter worm, summer grass’, in reference to its dual origins and transformation from larva to stroma. It is widely believed that the medicinal effect of the collective entity is in part due to its dual nature, and is therefore compatible with the Chinese philosophical concept of yin and yang. Its medicinal properties have been recorded for at least 500 years (Winkler 2005) and are reputed to have been discovered through observation of increased vigour of yaks grazing in the grasslands in which O. sinensis grows.

The primary use of yartsa guenbub in traditional Oriental medicine is as a tonic to aid in recovery from illness or disease, to build up strength and restore energy (Pegler et al. 1994; Jiang and Yao 2002). More recently it has been cited as possessing a range of more specific therapeutic properties, including action against asthma and bronchial inflammation (Kuo et al. 2001), cure of renal complaints (Guo et al. 1999) and stimulation of the immune system (Kuo et al. 2005). Perhaps inevitably, it has been ascribed aphrodisiac properties (it is referred to as ‘Himalayan Viagra’ in numerous websites devoted to alternative medicine) and it has even been claimed to be the secret of eternal youth (see Paterson 2008). Vinning and Tobgay (2004) provide what they refer to as a ‘near miraculous’ listing of what yartsa guenbub can address, including:

  • Improvement of respiratory functions.

  • Increase of cellular oxygen absorption.

  • Improvement of the functioning of the heart.

  • Improvement of cholesterol balance, increasing HDL cholesterol and lowering triglyceride cholesterol.

  • Promotion of DNA repair.

  • Improvement of liver function, helping to combat hepatis and cirrhosis.

  • Reduction of tumour size.

  • Assistance to the immune system by increasing NK cell activity.

  • Anti-aging through stimulation of hormone production and inhibition of monoamine oxidase enzymes.

  • Protection against free radical damage.

  • Combat of sexual dysfunction.

Due to the cultural importance placed in the dual nature of yartsa guenbub, it is used in China and surrounding countries largely in its original form—i.e. as the dried stroma emerging from the caterpillar cadaver. The quality of the product is also of paramount importance, with the highest prices paid for immature specimens with the insect component still intact and much lower amounts paid for broken or otherwise damaged material. There is a market for pharmaceutical and neutraceutical products (pills, capsules, tonics etc. often combined with other therapeutic substances) derived from Ophiocordyceps sinensis in Western alternative medicine, but these are less popular in the Orient and we are not convinced that all of them contain the genuine product. Some are marketed as derived from cultures (with the benefits to conservation stressed), but in our experience O. sinensis grows very slowly in culture and is unlikely to prove commercially viable in this form. In China, research into artificial rearing of yartsa guenbub by inoculating farmed Thitarodes larvae with O. sinensis spores has been pursued for a number of years (Winkler 2005). However, the results have not been highly publicized and if successful there has not been an obvious effect on market prices.

Yartsa guenbub is traditionally used by boiling the whole (dual) organism in water or soup, and then drinking the resulting liquid (Pegler et al. 1994). It may also be prepared in a more elaborate (and expensive) manner by as a stuffing in roast or boiled duck. In Bhutan, it is sometimes taken in a hot soup derived from home-distilled spirits which must compromise any therapeutic benefit, although the effect is undeniably warming when taken at high altitude.

It is appropriate to question some of the more extreme claims of health benefits of yartsa guenbub, especially those promoted by less scrupulous suppliers on the Internet. At least until comprehensive medical trials can be completed, we would not wish to deny more general therapeutic effects within the traditional Oriental medical system. Whether or not the claimed therapeutic benefits of yartsa guenbub can be substantiated, demand is such that conservation and sustainability of supply is a legitimate concern for the Himalayan nations.


Following its recognition as a therapeutic substance, the financial value of yartsa guenbub quickly appreciated, costing four times its weight in silver in the early eighteenth century (Du Halde 1736), and for many years it was only legally accessible by the Chinese nobility. Its economic importance persisted over the years with reports of yartsa guenbub being used as currency and being sold in exchange for luxury goods (e.g. Coales 1919; Bailey 1945). Prices were depressed during the Cultural Revolution and after, but by the early 1980s yartsa guenbub was traded in China for 800–1,000 RMB kg−1 (Winkler 2005). The true equivalent is difficult to compute due to the lack of free currency exchange at the time, but the sum probably equates to around US $500.

Since then, a number of factors have resulted in a very substantial increase in prices. Increasing interest by Western consumers in traditional Oriental medicine has led to an overall increase in demand, and this was stimulated greatly in 1993 by reports that Chinese athletes who broke a series of world records in distance events used yartsa guenbub as a neutraceutical as part of their training regime (Pegler et al. 1994; Steinkraus and Whitfield 1994). The outbreaks of Severe Acute Respiratory Syndrome (SARS) in 2003 and more recently of avian flu have also led to substantial increase in demand. The increase in affluence in China has also had a major impact, especially as yartsa guenbub is considered as a traditional gift from those with the necessary financial resources.

Official production figures for Tibet between 1999 and 2004 range between about 35 and 50 t annually, with figures for China as a whole estimated at between 100 and 200 t year−1 (Winkler 2007). Current prices vary widely depending on supply and quality, but the average wholesale value for yartsa guenbub re-exported from Hong Kong between 2000 and 2003 was around US $900 kg−1, based on “official” market volumes ranging between 24 t in 2000 to 13 t in 2,003 (Vinning and Tobgay 2004). In 2004 wholesale prices rose to around US $4,500 kg−1 (Winkler 2008), an increase of about 350% since 1997 after inflation is taken into account. In 2007 Chinese retail herbalists in Bangkok were selling yartsa guenbub from the 2006 harvest for up to US $17,000 kg−1, representing an approximately five times increase on prices paid to collectors in Bhutan the same year. Even higher prices—up to US $32,000 kg−1—were being charged for high-grade Tibetan yartsa guenbub in late 2006 in the coastal cities of China (including Hong Kong) and San Francisco (Winkler 2008).

Ophiocordyceps exploitation in Bhutan

The regulatory system

Yartsa guenbub has been a resource for Bhutanese traditional medicine for many years, although written records are sparse. Bhutan’s Biodiversity Action Plan was published in 2002. Ophiocordyceps sinensis was one of only seven non-animal species listed in Schedule 1 of the Forest and Nature Conservation Act of Bhutan, and anti-poaching measures were described (Anon 2002). Direct protection of the species is problematic as even in protected areas (a large proportion of the Ophiocordyceps habitat in montane Bhutan is within National Parks and other locations subject to conservation legislation) the number of foresters, park guards etc. is inadequate to prevent unauthorized harvest. Many of the harvest sites are close to the border with Tibet, which is easy to cross clandestinely in many places, and while there is also security support from the Royal Bhutan Army, the high value of the crop is clearly tempting to many.

Conservation measures were reviewed by the Royal Government, following a consultancy visit to the upper Soe Valley in the Jigme Dorji National Park (Hywel-Jones and Tshitila 2002). At that time, the collection ban appeared to be largely self-enforced by the indigenous peoples, but it was clear that villagers had a detailed knowledge of the distribution of yartsa guenbub in the hills. They were also well aware of the extent to which poachers entered illegally from Tibet to collect, but felt that there were few incentives to cooperate with the national park staff to enforce the ban. This was compounded by the fact that many of the families present legitimately in the National Park are intermarried with people from across the border, and are dependent on them for local trade.

The 2002 season for yartsa guenbub was delayed due to abnormally cold weather, but even so more than 20 non-national collectors were apprehended with over 40 kg of stromata, representing perhaps 100,000 individual units. A significantly higher amount must have been successfully smuggled out of Bhutan to make the enterprise worthwhile. In view of the likely level of unauthorized harvest, the consultancy report recommended a partial lifting of the collection ban in an attempt to achieve self-policing of the crop by the local villagers, who would then have a greater incentive to discourage external poaching.

The law was changed in 2004 to allow limited collection of yartsa guenbub by the yak herders whose herds traditionally graze in the pastures where the fungus occurs. Various measures were put in place to restrict the overall harvest, including a ban on collection except during the month of June, and a stipulation that only one member of each household was allowed to collect. These measures were considered to be largely successful, and have been maintained since with some modifications.

A key rationale behind the regulation on harvest dates was focused on the fact that immature specimens with the caterpillar cadaver still intact fetch much higher prices than mature stromata, where the nutritional reserves within the caterpillar exoskeleton have been exhausted. By that stage the exoskeleton is effectively empty of contents and usually breaks up when the yartsa guenbub is disturbed during the harvest process. The fungus fruit body in situ is very inconspicuous, and it is likely that a proportion of specimens will remain even after quite intensive harvest. This policy allows collectors to remove immature specimens and maximize their income, while protecting mature fruit bodies to allow spore dispersal. In the long term this acute selection pressure may favour genotypes that emerge later in the season, but must represent a least-worst scenario bearing in mind the difficulties of policing.

The regulatory system has been under annual review since 2004 through stakeholder workshops and consultation with village representatives. There were some minor changes to the time period for harvest to allow for later seasons. In 2008 the law governing the number of family members able to collect was relaxed, and in some cases decision-making was decentralized to local level. This had mixed success, with some indications that local leaders were adopting less stringent controls on collectors. A stakeholder workshop in July 2008 recommended a return to somewhat more tightly controlled system pending development of a locally focused natural resource management system, but with an extended collection period allowing harvest at any time in the season until the end of June.

The auction process

In Bhutan, yartsa guenbub can only be legally sold by the collectors at Government-administered auctions (Fig. 3). This brings an element of competition into the buying process and provides a degree of protection to the harvesters, who are mostly poorly educated and have little concept of the ultimate retail value of their crop. The auctions are held in various localities throughout northern Bhutan, at roadheads etc. which are accessible to buyers and relatively convenient for harvesters. The regulations state that only individual licensed collectors may sell yartsa guenbub at auction, though the quantities offered for sale range from a few units weighing 25–50 g to 1 kg or more, and it seems difficult to believe that those with such large quantities have been so fortunate in their searching.

Fig. 3
figure 3

Yartsa guenbub auction at Dodena, north-western Bhutan. a Yak herders waiting for their lots to be auctioned; a, b collection of good-quality yartsa guenbub ready for sale

There are indications that unauthorized sale of yartsa guenbub occurs outside of the auction system, by legitimate collectors to buyers from the north and also by unauthorized collectors to others with licences who can gain access to the auctions. We heard claims that the quantities offered for sale at auction from year to year varied depending on the relative prices being obtained in Bhutan compared with Tibet. In 2008 the large quantity on sale legitimately in Bhutan was considered in part due to the enhanced border security adopted by the Chinese authorities, to guard against possible Tibetan demonstrations relating to the Olympic Games in Beijing.

The auctions are highly colourful affairs, with the sellers looking forward to substantial financial gain and the buyers hoping to restrict the prices. Allegations of collusion to distort the market rates are made on both sides. Currently, the buyers must be Bhutanese nationals, who will either sell on what they buy internationally or feed the local market. There is one company at present linked to the Institute of Traditional Medicine Studies in Thimphu, which manufactures a health product in tablet form for local consumption from a mixture of yartsa guenbub and other medicinal herbs.

There were 13 separate auctions in 2008, with considerable variation in quantities sold and numbers of buyers and sellers. One auction attracted only five sellers with 125 g of yartsa guenbub changing hands, but over 100 sellers attended at the majority of events and at one auction 585 were present. Current regulations state that a seller can refuse the price offered at auction and offer his/her harvest at a second event, so enumerating the total number of collectors is difficult. The number of buyers ranged between seven and 27. The Royal Government places a levy on sale of yartsa guenbub at auction (paid by the buyers), which pays for the auction administration and also contributes towards environmental protection. The formula for its calculation has changed slightly over the past 5 years, but in 2008 the levy amounted to about 4.9% of total revenues.

The yartsa guenbub on sale by individual herders is often combined into larger lots of equivalent quality to speed up the auction process. Quality of the product is of paramount importance, with the best collections fetching several times the price of the lower-quality lots. Key features include the state of maturity (those with intact caterpillar cadavers fetching much more), the correct state of desiccation, the overall presentation (well cleaned, no broken pieces, bright colour, straight stromata etc.) and the smell. This last feature is difficult to assess by the non-afficionado but much of the crop is dried over wood smoke which is unattractive to the buyers.

Impact on local economies

Currently in Bhutan, yartsa guenbub harvesters must be ordinarily resident in the collection areas during the summer months. Almost all are yak herders operating a transhumance system of husbandry, with their animals grazing at low altitude during the winter months and moving up the valleys as the season develops. In some cases herders look after animals belong to others, or that are owned by local monasteries etc. Before the opportunities to collect yartsa guenbub presented themselves, almost their only income was the sale of yak products (milk or cheese) or slaughtering an animal in the autumn to sell for meat. Most own only a small number of animals, with a single yak fetching perhaps US $200. Other modest income is derived from hiring yaks (or more usually horses) as pack animals, both to tourists on trek and to the local market. In many cases goods are obtained by barter.

Yartsa guenbub harvest therefore provides a far greater income compared to their traditional activities for most yak herders. In 2008 in Bumdeling, for example, a set of 40 individual collectors received a total of nearly US $30,000 for their harvest (Table 1). 75% of them received at least the price of a yak, and 25% of them more than $1,000. One individual was paid more than $US2500—almost the annual salary of a graduate teacher. The income does represent the outcome of very considerable physical hardship. Similarly dramatic changes in the local economy have been reported for Tibet by Winkler (2008), where yartsa guenbub now accounts for between 50 and 80% of overall rural income in regions where it grows, and contributing at least US $225 million to the Tibetan GDP in 2004.

Table 1 Income obtained by 40 yartsa guenbub sellers at the Government auction in Bumdeling, 2008

Interviews with the sellers emphasize the life-changing nature of the new crop: a number expressed satisfaction that they would not need to sell a yak that year, and money was available for school fees and luxuries such as solar panels and mobile phones. Not surprisingly, this new affluence has been noted by other sectors of Bhutanese society, and there are calls for yartsa guenbub collection licences to be more widely available.

The long-term impact of yartsa guenbub harvest is very difficult to predict. Currently the prices obtained are attractive to collectors. Laws of supply and demand mean that overall income levels are buffered to some degree; in 2008 in Bhutan the quantities offered for sale were large but the prices obtained were substantially lower than the previous year (see Table 2). In the long term this must be good for the harvesters, though at the auctions this year some found it difficult to understand that their crop was less valuable that year than they anticipated.

Table 2 Summary of quantities sold of yartsa guenbub and prices obtained at auction in Bhutan, 2004–2008

There are at least three major potential factors that will impinge on prices in the future; the sustainability of supply, the health of the global economy (and especially the economy of China, by far the biggest market), and the availability and acceptability of farmed or manufactured alternatives. Further factors may also become significant, especially global warming. This may well be detrimental to yartsa guenbub populations as their environment becomes drier due to reduction in snow fall and suitable habitats shrink in size with a shift to higher altitudes.

At a more local level, some of the yak farmers interviewed was considering expanding their herds, but many have been buying land further down in the valleys and could give up yak husbandry in the future. Such decisions would certainly be influenced by possible changes in policy on entitlements to collect yartsa guenbub.

Environmental impact

Harvest of yartsa guenbub in Bhutan does have environmental consequences, although at present the impact is not substantially greater than that caused by yak farming. The montane grasslands where yartsa guenbub occurs are also occupied by other herbivorous mammals, including marmots and other small rodents that eat plant shoots and roots, and larger mammals such as the bharal (blue sheep) for which conservation measures are in place. Bharal are a principal food source for the snow leopard, leading to further interest in their preservation.

The impact of yak grazing on the environment is poorly understood. In some areas (particularly on migration routes) over-grazing has caused significant erosion, but in many areas it seems that the populations do not cause a major impact on the natural environment. Preliminary experiments on grazing exclosure (data not supplied) do indicate that yak grazing may have a significant effect on vegetation composition. The impact of yak grazing on yartsa guenbub production is also poorly understood. Direct fungivory by yaks is probably insignificant (despite traditional claims) and their absence might well lead to increases in consumption by other mammals. There may also be indirect effects. Changes in vegetation composition might well impact on Thitarodes caterpillar populations, though the evidence we have to date suggests that they are not strongly specific in their feeding habits. Substantial reductions in moth populations would certainly be detrimental to their fungal parasite. It is possible that yak grazing could be beneficial on balance to yartsa guenbub production, as the shorter turf may tend to allow more efficient dispersal of Ophiocordyceps sinensis spores. Current knowledge suggests a strong positive correlation between yartsa guenbub occurrence and yak pasture, although areas not grazed by yaks have not been fully surveyed.

Human interference has direct impacts on yartsa guenbub and its ecology, quite apart from its removal from the ecosystem. Stromata are typically removed from the ground by digging around them with a small spike or knife, which may leave holes in the turf. This is an especial problem with unauthorized collection, as fruit bodies are removed as rapidly as possible to reduce the risk of being caught poaching. Fuelwood gathering is also a major concern, both by removal of wood from the forests near the tree line, and by cutting dwarf rhododendron from the sites themselves. Increased human presence would tend to deter the more charismatic elements of biodiversity. The harvest areas frequently become polluted with plastic, wrappers, tins, cigarette stubs etc. left by the collectors, especially those that are intruders. The effect on the moth hosts is difficult to quantify but they fly late in the season when the weather is often poor and any remaining yartsa guenbub is of poor quality, so significant human-mediated deleterious effects are unlikely.

The biggest risk to the environment in the future of yartsa guenbub collection must be if the number of collectors continues to increase, and this of course is likely to impact in a major way on sustainability of harvest. That risk is likely to be exacerbated if prices continue to increase.

Sustainability of harvest

Yartsa guenbub has been collected in the region for centuries and there is no immediate concern that either the parasite or its host moth is seriously threatened. It is difficult to access accurate quantitative data for Bhutan on historical collection levels, but the rapidly increasing market prices in the past 10 years have undoubtedly led to new conservation pressures. Ophiocordyceps sinensis stromata are small and inconspicuous and grow in highly inaccessible places, so it is difficult to imagine that the species will be rendered extinct—at least solely as a result of human intervention. It is entirely possible, however, that population levels will be reduced to the level where harvest is uneconomic, with substantial deleterious effects on the local economy. In turn this would be likely to lead to yak husbandry also becoming uneconomic, with the farmers moving the the lower valleys and the cities in search of work. As the farmers provide most of the transport for trekking firms, the tourist industry would also be badly affected.

The Royal Government of Bhutan has already put a regulatory system in place (described above) to provide overall control of the yartsa guenbub market, and this will be reviewed on an annual basis. A long-term monitoring programme has also been put in place at a research site at around 4,700 m altitude at Namna, north-western Bhutan, and the first 3 years’ data are summarized in Table 3 and Fig. 4. Two further sites have been surveyed, one in north central Bhutan in the Bumthang valley and the other in the east of the country in the Bumdeling Wildlife Sanctuary, and it is hoped that these will form a research network in the future. The programme at Namna involves detailed study of a series of 20 × 2 m transects, recording incidence, phenology and development. Linked studies have addressed the moth populations, looking at larval incidence and potential instar number, possible food plants and flight period, as well as the identity of the moth species. A separate paper describing the moths found at Namna is in preparation. The research site is under the guardianship of a local yak herder who has shown great interest in the research. There is no practical means of protecting the site from poachers, and the research plots were undisturbed during the first 2 years’ survey. In 2008 some of the transects were interfered with which caused loss of some data, but overall population levels could still be assessed. The population density seems broadly similar to that observed elsewhere in the Himalayas (Winkler 2005).

Table 3 Summary of Ophiocordyceps sinensis population levels (stroma numbers) at the Royal government of Bhutan’s research site at Namna
Fig. 4
figure 4

Phenology of Ophiocordyceps sinensis at the Namna permanent study site, NW Bhutan from 2006 to 2008

There has been substantial variation in population levels, both within transects in specific years, and in overall population levels between years. This reflects the highly patchy distribution of Ophiocordyceps sinensis, which will be subject to natural fluctuations in population levels of both fungus and host. Populations are also likely to be affected by environmental conditions, including winter and summer temperatures and level and seasonality of precipitation. The population studies will need to continue for a number of further years before natural trends can be identified, allowing the data to be used to inform decisions on future harvest levels and regulation of collection.

Various consciousness-raising activities have also taken place to emphasize the importance of sustainable harvest. These have included a number of stakeholder workshops aimed at different levels of society, including the gups (village elders), foresters and staff of the Nature Conservation Division, and the members of parliament, established for the first time in early 2008. The message has also been disseminated to many of the collectors directly, primarily at the auctions. Due doubtless to the nation’s Buddhist heritage which emphasizes respect for the environment, there have been few dissenting voices in response to the sustainability message, but the lure of cash in the short term is great and it will be important to reinforce the concept regularly in future.

In the long term, the only viable way of ensuring sustainability of wild yartsa guenbub harvest is through locally focused natural resource management, with the villagers making their own informed decisions about collection policy. There has already been successful delegation of similar decision-making powers to the local level in Bhutan (e.g. Gurung et al. 2006). A pilot programme was promoted to the people of Lingshi and its satellite villages (Northwest Bhutan) in 2007 but met initially with limited success. The major concern was that the villagers have only very limited control over unauthorized collection by outsiders, and it was felt that any restrictions placed on collection by the villagers themselves would be negated by poachers. The solution would seem to be a larger- and longer-term programme with more effective coordination between local stakeholders and regulatory authorities (National Park staff, military patrols etc.). Yartsa guenbub has now been included into a community based natural resource management project funded by IDRC, so there is some prospect of further success.


Yartsa guenbub looks set to provide substantial long-term income to the indigenous peoples of the Tibetan plateau, but unless harvest levels are sustainable local economies will suffer major decline in the medium to long term. This could have economic implications on a national scale for Bhutan, due to the current income levels and the difficulties of diversification. This paper documents some of the measures that have been put into place to ensure sustainability of production, but continued vigilance and restraint will be needed—at least while reliable baseline population data are gathered.

Farming of yartsa guenbub might provide a long-term solution to sustainability of supply, through cultivation of Thitarodes larvae and artificial inoculation with Ophiocordyceps sinensis spores. There has apparently been research in China for some years on this subject (Winkler 2005), but though success has been claimed, we are not aware of any positive research results published in peer-reviewed journals. There are major obstacles to developing a farming system, due to the intricacies of host-parasite relations and the extreme environment in which the two species naturally occur. A “factory-farming” approach would have devastating consequences on local livelihoods, although if it was possible to develop small-scale, low-impact farming enterprises in the natural habitat, these might provide better long-term environmental protection for the high Himalayan grassland ecosystems. “Seeding” of Ophiocordyceps habitats with Thitarodes eggs has been attempted, though results have not been publicized (Stone 2008); this approach could increase population density without causing environmental or socioeconomic damage.

A regional approach to Ophiocordyceps conservation must be a high priority, to share best practice, establish common sustainability targets, coordinate regulatory policy and reduce cross-border activity. This has been recognized as a high priority by the Royal Government of Bhutan, and will be taken forward as funding opportunities are identified.


  1. As this paper focuses on Bhutan, the product will be referred to using the Dzongkha term.


  • Andersson S, Rydell J, Svensson MGE (1998) Light, predation and the lekking behaviour of the ghost swift Hepialus humuli (L.) (Lepidoptera, Hepialidae). Proc R Soc Lond B Biol Sci 265:1345–1351. doi:10.1098/rspb.1998.0440

    Article  Google Scholar 

  • Anon (2002) Biodiversity Action Plan for Bhutan 2002. Ministry of Agriculture, Royal Government of Bhutan, Thimphu

    Google Scholar 

  • Bailey FM (1945) China-Tibet-Assam: a Journey, 1911. Jonathan Cape, London

    Google Scholar 

  • Berkeley MJ (1843) On some entomogenous Sphaeriae. Hooker’s Lond J Bot 2:205–211

    Google Scholar 

  • Borel JF (2002) History of the discovery of cyclosporin and of its early pharmacological development. Wien Klin Wochenschr 114:433–437

    PubMed  CAS  Google Scholar 

  • Chen Y, Zhang Y-P, Yang Y, Yang D (1999) Genetic diversity and taxonomic implication of Cordyceps sinensis as revealed by RAPD markers. Biochem Genet 37:201–213. doi:10.1023/A:1018738706123

    PubMed  Article  CAS  Google Scholar 

  • Chen SJ, Yin DH, Zhong GY, Huang TF (2002) Study on the biology of adult parasite of Cordyceps sinensis, Hepialus biruensis. Zhongguo Zhong Yao Za Zhi 27:893–895

    PubMed  Google Scholar 

  • Coales O (1919) Economic notes on eastern Tibet. Geogr J 54:242–247. doi:10.2307/1779942

    Article  Google Scholar 

  • Devkota S (2006) Yarsagumba (Cordyceps sinensis (Berk.) Sacc.); traditional utilization in Dolpa District, western Nepal. Our Nat 4:48–52

    Google Scholar 

  • Du Halde P (1736) The General History of China, vol 4. John Watts, London

    Google Scholar 

  • Guo HP, Yang ZM (1999) Advances in pharmacological study of Cordyceps sinensis. Zhong Cao Yao 30:231–233

    Google Scholar 

  • Gurung TR, Bousquet F, Trébuil G (2006) Companion modelling, conflict resolution and institution building: sharing irrigation water in the Lingmuteychu watershed, Bhutan. Ecol Society 11(2):49

    Google Scholar 

  • Hywel-Jones NL, Tshitila (2002) Cordyceps sinensis in Bhutan. Unpublished consultancy report to the Ministry of Agriculture, Royal Government of Bhutan

  • Jiang Y, Yao Y-J (2002) Names related to Cordyceps sinensis anamorph. Mycotaxon 84:245–254

    Google Scholar 

  • Kuo YC, Tsai WJ, Wang JY, Chang SC, Lin CY, Shiao MS (2001) Regulation of bronchoalveolar lavage fluids cell function by the immunomodulatory agents from Cordyceps sinensis. Life Sci 68:1067–1082. doi:10.1016/S0024-3205(00)01011-0

    PubMed  Article  CAS  Google Scholar 

  • Kuo CF, Chen CC, Luo YH, Huang RY, Chuang WJ, Sheu CC, Lin YS (2005) Cordyceps sinensis mycelium protects mice from group A streptococcal infection. J Med Microbiol 54:795–802. doi:10.1099/jmm.0.45704-0

    PubMed  Article  Google Scholar 

  • Li QS, Zeng W, Yin DH, Huang TF (1999) A preliminary study on alternation of generations of Cordyceps sinensis. Zhongguo Zhong Yao Za Zhi 23:210–212

    Google Scholar 

  • Nabokov V (1991) The gift. Vintage International, New York

    Google Scholar 

  • Namgyel P, Tshitila (2003) Rare, endangered, over-exploitation and extinction of plant species? Putting Cordyceps—a high value medicinal plant-to test. Council of Research and Extension and Bhutan Trust Fund for Environmental Conservation, Thimphu

    Google Scholar 

  • Nielsen ES, Robinson GS, Wagner DL (2000) Ghost-moths of the world: a global inventory and bibliography of the Exoporia (Mnesarchaeoidea and Hepialoidea) (Lepidoptera). J Nat Hist 34:823–878. doi:10.1080/002229300299282

    Article  Google Scholar 

  • Oberthür C (1909) Études de Lépidoptérologie Comparée. Part 3 Privately published, Rennes

  • Paterson RRM (2008) Cordyceps—a traditional Chinese medicine and another fungal therapeutic biofactory? Phytochemistry 69:1469–1495. doi:10.1016/j.phytochem.2008.01.027

    PubMed  Article  CAS  Google Scholar 

  • Pegler DN, Yao Y-J, Li Y (1994) The Chinese ‘caterpillar fungus’. Mycologist 8:3–5

    Article  Google Scholar 

  • Saccardo PA (1878) Enumeratio Pyrenomycetum Hypocreaceorum hucusque cognitorum systemate carpologico dispositorum. Michelia 1:277–325

    Google Scholar 

  • Schardl CL, Panaccione DG, Tudzynski P (2006) Ergot alkaloids–biology and molecular biology. Alkaloids Chem Biol 63:45–86. doi:10.1016/S1099-4831(06)63002-2

    PubMed  Article  CAS  Google Scholar 

  • Sharma S (2004) Trade of Cordyceps sinensis from high altitudes of the Indian Himalaya: conservation and biotechnological priorities. Curr Sci 86:1614–1619

    Google Scholar 

  • Steinkraus DC, Whitfield JB (1994) Chinese caterpillar fungus and world record runners. Am Entomol 40:235–239

    Google Scholar 

  • Stensrud Ø, Schumacher T, Shalchian-Tabrizi K, Svegården IB, Kauserud H (2007) Accelerated nrDNA evolution and profound AT bias in the medicinal fungus Cordyceps sinensis. Mycol Res 111:409–415. doi:10.1016/j.mycres.2007.01.015

    PubMed  Article  CAS  Google Scholar 

  • Stone R (2008) Last stand for the body snatcher of the Himalayas? Science 322:1182. doi:10.1126/science.322.5905.1182

    PubMed  Article  CAS  Google Scholar 

  • Sung G-H, Hywel-Jones NL, Sung J-M, Luangsa-ard JJ, Shrestha B, Spatafora JW (2007) Phylogenetic classification of Cordyceps and the clavicipitaceous fungi. Stud Mycol 57:1–63

    Article  Google Scholar 

  • Viette PEL (1968) Contribution à l’études des Hepialidae (36ème note): lepidoptera Hepialidae du Népal. Khumbu Himal 3:128–133

    Google Scholar 

  • Vinning G, Tobgay S (2004) Cordyceps: a Market Analysis. Ministry of Agriculture, Thimphu

    Google Scholar 

  • Winkler D (2005) Yartsa Gunbu—Cordyceps sinensis. Economy, ecology and ethno-mycology of a fungus endemic to the Tibetan Plateau. Mem Soc Ital Sci Nat Mus Civico di Storia Nat Milano 33(1):69–85

    Google Scholar 

  • Winkler D (2007) Yartsa Gunbu (Cordyceps sinensis (Berk.) Sacc.). Present and historic relevance in Tibet. Abstracts. Fourth International Medicinal Mushroom Conference, Ljubljana

    Google Scholar 

  • Winkler D (2008) Yartsa gunbu (Cordyceps sinensis) and the fungal commodification of Tibet’s rural economy. Econ Bot 62:291–305

    Google Scholar 

  • Xing XK, Guo SX (2008) The structure and histochemistry of sclerotia of Ophiocordyceps sinensis. Mycologia 100:616–625. doi:10.3852/07-007R2

    PubMed  Article  CAS  Google Scholar 

  • Zang M, Kinjo N (1998) Notes on the alpine Cordyceps of China and nearby nations. Mycotaxon 66:215–229

    Google Scholar 

  • Zhu HF, Wang LY, Han HX (2004) Lepidoptera: Hepialidae, Epiplemidae, Insecta. Fauna Sinica, vol 38. Science Press, Beijing

    Google Scholar 

  • Zimmer DE (2002) Chinese rhubarb and caterpillars. Unpublished paper read at the International Vladimir Nabokov Symposium, St Petersburg, 18 July 2002; accessible at

  • Zimmer DE (2006) Nabokov reist im Traum in das Innere Asiens. Rowohlt Verlag, Reinbek

    Google Scholar 

Download references


The research described in this paper was funded by a grant from the UK Government’s Darwin Initiative, to a partnership between CABI, CoRRB (Ministry of Agriculture, Bhutan) and BIOTEC (National Centre for Genetic Engineering and Biotechnology, Thailand). We received assistance from numerous individuals, agencies and institutions in Bhutan, including staff of the Nature Conservation Division, the National Biodiversity Centre, the Agricultural Marketing Service and the Institute of Traditional Medicine. We would like to pay particular tribute to the CoRRB field research staff (especially Kuenzang Dhendup, Dawa Tshering and T.B. Rai), who spent many weeks at high altitude in often arduous conditions to gather baseline data on Ophiocordyceps sinensis and its host moths. Information on sales at auction was provided by Sonam Tobgay (AMS) and Dophu Drupka and Sonam Wangmo (CoRRB).

Author information

Authors and Affiliations


Corresponding author

Correspondence to Paul F. Cannon.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Cannon, P.F., Hywel-Jones, N.L., Maczey, N. et al. Steps towards sustainable harvest of Ophiocordyceps sinensis in Bhutan. Biodivers Conserv 18, 2263–2281 (2009).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Ophiocordyceps sinensis
  • Yartsa gunbu
  • Yartsa guenbub
  • Thitarodes
  • Sustainable harvest
  • Traditional Oriental medicine
  • Bhutan