Cultivating China’s Cinchona: The Local Developmental State, Global Botanic Networks and Cinchona Cultivation in Yunnan, 1930s–1940s

Summary This article reconstructs the history of China’s first successful cinchona cultivation programme in Hekou, Yunnan province from the 1930s to 1940s during the Nationalist era (1928–49). I argue that the Hekou programme was initiated by the Yunnan ‘local developmental state’ to control endemic malaria and achieve quinine self-sufficiency. It was expanded during the Sino-Japanese War (1937–45) as part of the national defence project in order to develop Yunnan’s malaria-ridden southwest frontier to provide more resources for the war, as well as to solve broader wartime epidemic crises in southwest China. A closer examination also indicates that the development of the Hekou programme was closely intertwined with global networks of cinchona cultivation and international politics.

empire', 4 and cinchona cultivation as an 'imperial science' 5 or a 'colonial botany' 6 for Euro-American and Japanese imperial expansion and colonial rule in their malaria-ridden tropical colonies.
As its Western counterparts, the Hekou cinchona cultivation programme was not a pure scientific agricultural experiment, but it served for more complicated political and economic purposes rather than a simple colonial or imperial project: it was initiated in the early 1930s by Zhang Banghan and his colleagues, in order to control endemic malaria and save national revenue for building Yunnan governor and warlord Long Yun's (1884Yun's ( -1962 'local developmental state', which resembled the technocratic central government of Nationalist China (1928-49). After the outbreak of the Sino-Japanese War , the Hekou programme was incorporated into the Chinese national defence project. In a certain form of 'internal colonialism', it was then advanced to develop Yunnan's semi-autonomous and malaria-ridden southwest frontiers to provide more resources for wartime needs. For another, together with other state-led bioprospecting programmes, it was also promoted by the central government to solve broader wartime epidemic crisis in Southwest China. In addition, the war played a key role in destabilising the Hekou programme. A closer examination on its integrations into Yunnan's local state building and Chinese national defence projects also indicates that the Hekou programme was intertwined with global networks of cinchona cultivation and international politics.

The Local Developmental State, Malaria and Yunnan Initiatives
Cinchona barks were brought to China by Jesuit missionaries in 1693 in order to cure Kangxi (r. 1661Kangxi (r. -1722, the second emperor of Qing China who suffered a serious malarial fever. For the next hundred years, the knowledge and application of cinchona barks gradually spread beyond the imperial court. 7 Since the mid-nineteenth century, isolated quinine pills had also been introduced and widely used in China along with European imperial expansions. 8 Recognising the medicinal and economic importance of the plant, some Chinese initiatives had attempted to cultivate cinchona in several southern Chinese provinces during the 1920s, but all of them had failed. It was not until 1935 that cinchona trees were successfully cultivated in Hekou County Experimental Farm for Tropical Plants, which was initiated by Zhang Banghan and his colleague Huang Riguang Medicine and Hygiene, 1937, 40, 187-88; Wang Jiming, 'Guanyu jinji'na chuanru woguo de jizai ', Zhonghua yishi zazhi, 1954, 6, 28-31;Marta Hanson, 'Jesuits andMedicine in the Kangxi Court (1662-1722) ', Pacific Rim Report, 2007, 43, 1-10. 8 For the history of quinine and its global spread, see Andreas-Holger Maehle, Drugs on Trial: Experimental Pharmacology and Therapeutic Innovation in the Eighteenth Century (Amsterdam and Atlanta: Editions Rodopi B. V., 1999), 223-309; James L. A. Webb, Jr., Humanity's Burden: A Global History of Malaria (New York: Cambridge University Press, 2009), 92-126. On the introduction and consumption of quinine in China in the nineteeth and twentieth centuries, see Yubin Shen, 'Malaria and Global Networks of Tropical Medicine in Modern China, 1919-1950' (unpublished PhD thesis, Georgetown University, 2017 The Yunnan Commission of Reconstruction led by Zhang Banghan was a crucial part of the provincial administration of warlord Long Yun's 'local developmental state'. The term 'developmental state' was originally used by Chalmers Johnson to describe the Japanese state's regulation and planning in Japan's economic development. 9 By applying this concept into modern Chinese history, scholars have already demonstrated that Nationalist China (1928-49) was an embryonic developmental state run by technocrats. 10 Following its founder Sun Yat-sen's (1866Yat-sen's ( -1925 developmentalist plans, 11 the Nationalist government appropriated the technocratic authority to legitimate its developmental policies in water control and national reconstruction. 12 Some recent studies have already moved to look at local state building carried out by provincial authorities, 13 such as warlord Chen Jitang  of Guangdong province in South China, who relied on agricultural experts to carry out his local developmental state project of the sugar industry. 14 Long Yun ruled Yunnan in a manner consistent with his Guangdong counterpart. After he took control of this province in 1927, Long Yun called for 'reconstructing a new Yunnan' and established several new and effective economic, military, educational and public health institutions, including the Yunnan Commission of Reconstruction, 15 and China's first provincial governmental agency specifically designed to combat endemic malaria, the Yunnan Anti-Malaria Commission (YAMC) in 1939. 16 Zhang Banghan was a lifelong disciple of Sun Yat-sen and a strong political supporter of Long Yun. Originally from Yunnan, Zhang joined Sun Yat-sen's Anti-Qing Revolutionary Alliance in 1906. He was then sent by Sun to study in Europe in 1909 and later received a Belgian diploma in Electrical Engineering. After the establishment of the 9 Chalmers Johnson, MITI and the Japanese Miracle: The Growth of Industrial Policy,1925-1975(Stanford: Stanford University Press, 1982 , 1927-1937(New York: Routledge, 2002 Nationalist regime in Nanjing in 1928, Zhang was appointed by Long Yun as the Provincial Commissioner of Reconstruction, a position he would hold until 1945. 17 During his tenure, Zhang avidly sought to realise Sun and Long's developmentalist ideas. In a special issue of the journal Chinese Reconstruction (Zhonguo jianshe) in 1934, Zhang elaborated his reconstruction plans for Yunnan, including promotion of modern farming and afforestation, as well as building modern transportation, hydraulic works and electricity systems. Considering the sub-tropical and tropical environmental conditions in south Yunnan, he proposed there should be farms for tropical cash crops, such as coffee, tung trees and plantains. 18 His interest in agriculture and forestry was not exceptional. During the early twentieth century, many intellectuals and officials believed that in order to save China, it would be necessary to modernise Chinese agriculture and afforestation using the science of forestry. 19 Since he did not know much about forestry himself, Zhang turned to Huang Riguang. Huang had graduated from the National Institute of Agriculture in Paris, and in 1924, he became a professor and chair of the experimental farm at the College of Agriculture of the newly established Guangdong University. The College of Agriculture, formerly an independent Guangdong Provincial Agricultural School, was incorporated into the university by Sun Yat-sen, who was convinced of the importance of modern agriculture, in order to build a strong focus on agriculture to ensure a better Chinese livelihood. 20 It soon became one of the leading agricultural colleges in China, especially famous for forestry. 21 In this particular college, Huang established himself as an expert on agriculture and tropical plants. 22 Invited by Zhang, Huang went to Yunnan in 1928 to serve as director of the Forest Section in the Commission of Reconstruction. 23 Both as technocrats trained in Europe, Zhang Banghan and Huang Riguang believed that by using modern scientific methods and rational management styles, they could more efficiently develop Yunnan. In the years that followed, they set up several institutes for the improvement of rice production, sericulture, tobacco, cotton, animal husbandry, tea, pest control and trees that could be grown for profit. They also established a provincial-wide system of tree farms for the purpose of afforesting Yunnan. 24 It was Huang who first persuaded Zhang to cultivate cinchona trees in Yunnan. 25 China, 1911-1937', Environmental History, 2003 After the death of Sun, the university was renamed as Sun Yat-sen University in 1926. Deng Zhiyi, 'Lun nongxueyuan zhi guoqu yu jianglai', Nong shen yuekan, 1933, 171, 1. and supplier in the world. He started experimenting on the college farm but unfortunately failed in his attempts. 26 Huang convinced Zhang of the importance of continuing his cinchona cultivation effort in Yunnan, using three major arguments: First, Yunnan, particularly its southwest frontier, had been a notoriously malaria-ridden region. At that time, over 10,000 patients died from malaria every year in Yunnan due to the shortage of quinine supplies. For example, Simao, a prosperous regional commercial centre in southwestern Yunnan, fell victim to a devastating malaria epidemic in 1919, lasting nearly 30 years. As a result, in 1927, its population had dropped from 76,800 to 24,106. After that, the population continued to decrease: in 1935, only 4,000 survivors lived there, and in 1953, Simao became little more than a ghost town, with only 1,092 residents. 27 Thus, cultivating cinchona (to produce quinine) would save people's lives, 28 provide healthier labourers for Long Yun's local developmental state and help strengthen his ruling legitimacy.
The second argument was saving national revenue. Malaria was endemic not only in Yunnan but also throughout China's southern provinces. Zhang and Huang pointed out that 8-10 tons of quinine pills were imported from overseas every year, resulting in a huge loss of government funds. What is worse, the quantities of imported quinine were only sufficient to meet the needs of less than one-third of the total malarial patients in China, but increasing the import threefold would have meant yet more expense. 29 According to another calculation in the 1930s, around 16 tons of quinine would have been required in order to treat all of the malarial sufferers in Yunnan. And throughout all of China, the total demand for cinchona bark, not to mention quinine products, was around 2 million jin (100 tons) every year, which would have amounted to a cost for the government of 150 million Chinese dollars (Table 1). 30 The idea of achieving autarky and building China's national economy was widely accepted by officials in their national reconstruction projects. 31 Zhang and Huang were among those technocrats who believed in these principles. They could not abide the fact that China was obliged to pay out huge sums for quinine imports, and thus, they planned to cultivate cinchona trees to save government expenditure, achieve quinine self- sufficiency, and possibly, in the process, gain considerable profits for the Chinese nation and the Yunnan local developmental state in the future. 32 The final reason had to do with the fact that Yunnan possessed a potentially favourable climate and environment for growing cinchona trees. Cinchona with its strict environmental requirements is perhaps one of the most difficult plants to cultivate. According to contemporary Western scientific experts, cinchona trees with a high proportion of quinine would only grow at latitudes from 10 north to 20 south, at altitudes from 1,200 to 2,000 m, with average temperatures from 18 C to 22 C, annual rainfall from 2.5 to 3.5 m, with little monthly variation and humus-rich soil. 33 Although Yunnan's latitudes were from 20 to 28 north, taking into consideration its particular tropical climate and soil conditions, Zhang and Huang believed that cinchona trees could be cultivated in the province's south and southwest malaria-ridden regions. 34

Global Botanical Networks and Cinchona Cultivation in Yunnan
When Zhang Banghan and Huang Riguang started their cinchona programme in Yunnan in the 1930s, the first difficult task for them was to acquire cinchona seeds. Educated in Europe, they recognised that there were already extensive global botanical networks of cinchona cultivation in place they could turn to.
Global botanical cinchona cultivation networks were initiated by the British and Dutch empires in the mid-nineteenth century, joined later by other imperial powers who also participated in cinchona cultivation for their own specific colonial agendas. 35 British colonial botanists and naturalists, such as Clements R. Markham and Richard Spruce, began to smuggle cinchona seeds out of South America and cultivated the plants in India in the 1850s, in order to safeguard British imperial enterprise in this malaria-ridden colony and to save huge expenditures on importing cinchona barks from South America. With support from the British Indian government and Kew Botanic Gardens, in a 'global network of exploration collection and systematisation of botanical knowledge', 36 cinchona was successfully cultivated in places like the Nilgiri Hills in British India in the 1860s. After that, through Kew Garden's 'global reach and its connections with state power' and its branch botanical gardens in various British colonies (including the Calcutta Botanic Garden in Bengal, which would play an important role in Yunnan's cinchona cultivation), many cinchona plantations were established in India, East Africa, the Caribbean, Ceylon (Sri Lanka) and Burma. 37 At almost the same time, in the 1850s, the Dutch government also started cinchona cultivation efforts in the Netherlands East Indies, particularly in Java, which boasted a 32 YPA, 77-23-76, 'Yunnan Hekou tuiguang zhongzhi jinji'na shu ji shiyan jihua shu, January 2, 1940'. 33 Lin Xixun, 'Yunnan zhi jinji'na ', Nong Shen, 1942, 225, 82- Geographical Journal, 1962, 128, 431-442; and Drayton, Nature's Government, 206-11. similar topography and climate to South America. Under the direction of colonial scientists such as Franz Junghuhn and K. W. van Gorkom and with the Dutch government's support, cinchona plantations in Java experienced rapid development. They began to enjoy huge economic profits, especially after 1872 when Charles Ledger's seeds (Cinchona ledgeriana) were widely planted in Java. 38 From 1890 onwards and up until 1940, the Netherlands East Indies gradually replaced British India and dominated the cinchona bark trade, by providing more than 90 per cent of the world's supply. 39 The French started their cinchona cultivation experiments in 1848 but soon faced failure. In 1869, seeds of cinchona trees were transported from Java to Saigon Botanic Garden in French Indochina. After several major failures in the five decades that followed, led by the famous bacteriologist Alexandre Yersin (1863-1943) of the Pasteur Institute in Saigon, cinchona trees were finally successfully cultivated in the 1920s in southern Vietnam. 40 The German cinchona project was initiated in 1900 in the colony Tanganyika of East Africa, while the Americans cultivated cinchona trees in the Philippines in 1927. 41 The Japanese Empire opened its own cinchona cultivation plan in the 1880s and succeeded in 1922 in its tropical colony Taiwan. 42 The Yunnan initiatives understood that the global networks were of an imperial and colonial nature, but still, they struggled to take advantage of them. At first they tried to get cinchona seeds from Java, but for reasons still not clear, the Dutch colonial government seemed to have placed strict restrictions on the exportation of cinchona seeds to the Chinese in the early 1930s. 43 2011, 18, 47-91. 43 According to Norman Taylor, it was not until 1937 that the Dutch colonial government began to embargo cinchona seeds, and before that many countries successfully purchased cinchona seeds from Java. See Norman Taylor Tropical Plants in Hekou County along the southwest borderland. Surprisingly, they grew very well, and the young plants soon began to leaf out. But the director of Hekou Farm, Zhang Jiliang, noticed that the leaves were completely different from the phyllotaxy of cinchona as described in textbooks. At that time, the famous American explorer and botanist Joseph Rock 46 (1884-1962) was travelling in Yunnan. Huang Riguang asked him to examine the plants. The results were very striking: these plants turned out to be ficus (banyan), not cinchona at all! There were two possible explanations for this mix-up: either the suppliers had given ficus seeds in the first place, or the cinchona seeds had been stolen and replaced with ficus mid-journey somewhere in Vietnam. Whatever the explanation, this further supports indication that obtaining cinchona seeds was still not an easy task.
In the winter of 1933, Handley-Derry once again purchased eight ounces of cinchona seeds for Yunnan initiatives from the Calcutta Botanical Garden. This time, he carefully packed and delivered them in secure diplomatic bags, and Yunnan project finally obtained authentic cinchona seeds. 47 Cinchona seeds from Calcutta were all brought to the Hekou Experimental Farm for Tropical Plants for cultivation. This farm was established in 1930 mainly for planting urban street trees, 48 and in 1933, it was reorganised for experimenting with tropical plants such as coconuts, bananas and coffee. 49 Zhang Banghan and his colleagues understood that the Hekou farm was not the perfect place to grow cinchona. The soil there was sandy loam, and its altitude was only 100 meters, which did not meet some of the critical requirements for cinchona cultivation. However, located on the border between Yunnan and Vietnam, it was the southernmost statecontrolled farm and the only available one that was located around the latitude of 20 north in Yunnan. What is more, Hekou had certain climatic advantages that might favour the growth of cinchona trees, such as an average temperature above 21 C. For these reasons, the Yunnan initiatives still considered it to be worth a try. 50 There was another possible advantage for cultivating cinchona trees at the Hekou Farm: the director and staff, trained as professional forestry experts, were strongly determined to transform Yunnan through means of modern agricultural science. The director, Zhang Jiliang, graduated from the School of Agriculture, Guangdong University, with a specialism in forestry. 51 Lin Yongxin, the deputy director, was also an alumnus from that college. As Lin recalled years later, 'to cultivate cinchona trees to save the patients' lives', they were 'resolute in their determination'. 52 Since 'the foreigners kept the knowledge of cinchona cultivation as secrets', they had little practical information or experience in cultivating this unfamiliar plant. 53 They attempted to sow cinchona seeds eight times, but the trees all failed to sprout. Zhang Banghan encouraged them by citing the example of the drug Salvarsan, experiments with which had failed 606 times before this magic bullet against syphilis was created. 54 On the ninth try, Zhang Jiliang and his staff in Hekou experimented with a technological innovation: they planted the seeds in boxes, and they finally germinated. Over 1,800 young plants were then successfully transplanted to the hills of the Hekou Farm. After 1936, they were even able to sow the seeds obtained from cinchona trees growing in the Hekou Farm itself. 55 This piece of news was soon broadcast around China, and Zhang Banghan and Hekou Farm became wellknown in China's scientific and medical circles. It was widely believed that following the Hekou model, the Chinese would continue 'acclimating cinchona trees to Chinese environments'. 56 In the next years, Yunnan continued to acquire cinchona seeds through the global botanical networks. In 1939, the French government presented them with cinchona seeds as gifts. 57 In 1940, the British consulate sent them seeds of red cinchona (Cinchona succirubra). 58 The American consulate offered 5.6 g of three different types of cinchona seeds in 1942. 59 By the 1940s, Yunnan had become part of the global network of cinchona cultivation.

Cultivating Cinchona for the War: Developing the Pusi Frontier and Bioprospecting Anti-Malarial Drugs
The Yunnan state-building technocrats were not satisfied with their minor success in Hekou. They drafted ambitious plans to cultivate millions of cinchona trees in the province's southwest region or the Pusi frontier, which seemed to boast more suitable land at higher altitudes. After the outbreak of the second Sino-Japanese war in 1937, two new motivations pushed them to put some of their plans into practice. The first one was developing Yunnan's Pusi frontier to supply resources and food for wartime needs. After the outbreak of the war, the Nationalist central government retreated to southwest China to ensure its survival. Over one million refugees from Japanese-occupied areas fled to Yunnan. To support the war and the huge population, it was critical to stimulate Yunnan's economy further to provide more food and resources. 60 As Long Yun claimed in a speech on the relationship between the war and developing Yunnan, the province boasted rich resources, such as water power and 'uncultivated wasteland', which could be better utilised to achieve final victory against the Japanese. 61 Among these 'fertile wastelands', Pusi was regarded as 'the perfect cultivable land for agriculture', 62 but it was a challenging mission to develop this 'disobedient and dangerous' frontier. indigenous regimes had maintained semi-autonomy away from the Chinese central government for hundreds of years. 63 Ever since the late eighteenth century, the Qing government had substantially expanded its state power by transforming some of these local chieftainships into direct administrations and encouraging move 'sophisticated' Han Chinese peasants into this area to cultivate the land. 64 However, the Pusi frontier was notorious for its serious malaria endemicity. As some scholars have pointed out, while the local ethnic Tai people were able to survive, having developed a hard-won resistance against malarial parasites, for most of the newly arrived Han Chinese immigrants, the disease had fatal consequences. This was the land where 'no Han could go for long'. 65 Such a long-term disease frontier environment had been maintained during the 1930s and 1940s, which posed a critical obstacle to the proposed cultivation plans of Long Yun's local developmental state. 66 The solution for eliminating malaria in Pusi frontier was to move more Han Chinese to reclaim land there during the Qing period. Republican intellectuals and officials held a similar view: '. . . there is a mutual waning and waxing relation between reclamation and malaria . . . if more people move in, and all wastelands are reclaimed, malaria endemicity will just go away'. 67 It was similar as the 'bonification' measure promoted by European malariologists in the late 1920s, which 'signified all work carried out with the object of making regions that are periodically or permanently marshy more healthy and more suitable for agriculture', as the drainage and filling of swamps could eliminate breeding places of the malarial mosquitoes. 68 After cinchona trees were successfully cultivated in Hekou, republican technocrats began to view cinchona tree plantations together with Han Chinese agricultural immigration as the best solution to develop the Pusi Frontier. 69  Hekou. 70 To collect more information about the Pusi frontier, and seek suitable places for planting cinchona trees there, Huang Riguang led the Investigating Team of the Pusi Frontier from November 1938 to February 1939, including Yuanjiang, Ning'er, Simao, Cheli, Fohai and Nanqiao Counties. During this trip, they left bottles of cinchona seeds and instructed local officials to try to cultivate cinchona trees there. The major success was achieved in Nanqiao. Zhang Jiliang brought cinchona seeds to Nanqiao Forest Farm, finding that its altitude of 1,400 meters was more suitable for cultivating cinchona, and thus a sub-programme was set up to cultivate cinchona there. 71 The role of cinchona cultivation in developing the Pusi frontier to some extent resembled Euro-American and Japanese counterparts in their tropical colonies, but more precisely, it served as 'a tool of internal colonialism' rather than a 'tool of empire'. As James Scott puts in his study on Zomia, 'the encounter between expansionary states and selfgoverning peoples' in this region was a 'cultural and administrative process of internal colonialism that characterizes the formation of most modern Western, nation states', which 'involved a botanical colonization in which the landscape was transformed by deforestation, drainage, irrigation, and levees to accommodate crops, settlement patterns, and systems of administration'. 72 The promotion of cinchona cultivation in the malaria-ridden Pusi frontier, together with Han Chinese immigrations during the wartime period, had undeniably shaped the 'internal colonial' nature of the Hekou programme.
Another important motivation was the central government's need for anti-malarial drugs, following its retreat to malaria-infested southwest China. Even though the Allies and international organisations provided huge amounts of quinine pills, the total demand could not be met. There was a nation-wide, desperate search for Chinese anti-malarial drugs that might serve as a substitute for quinine, otherwise known as 'bio-prospecting', 73 which also paralleled similar search projects undertaken elsewhere, such as in British India and post-colonial Africa. 74 The most famous case was the promotion of the Chinese herb changshan (Dichroa febrifuga). 75 After its efficacy against malaria was proved by Chinese scientists, the Nationalist central government set up several changshan cultivation projects in the 1940s. 76 Nevertheless, within this context of seeking anti-malarial drugs, even more attention was paid to China's own cinchona trees in Yunnan. As early as 1936, the Fourth Congress of the Chinese Medical Association had already proposed that the Nationalist central government promote cinchona cultivation in China's tropical provinces-Guangdong, Guangxi, Yunnan and Guizhou-in order to treat malaria-infested patients, to save revenue and to prepare for a possible world war. 77 In 1939, having heard the news about Hekou's successful cinchona cultivation, the Congress of the Chinese Medical Association sent an official letter to inquire into the details of the Yunnan initiative's success. The Yunnan Commission of Reconstruction saw it as a good opportunity to attract the central government's attention and ordered the Hekou Farm to prepare reports of its cultivation achievements for both the Congress of the Chinese Medical Association and the central government.
Hekou Farm also tried to gain more funding by proving that their cinchona bark contained a high percentage of quinine alkaloids and was thus effective in practical treatment. In November 1939, they sent several bark samples to the Army Factory of Medical Materials, and the reports indicated that these samples contained 6.54 per cent quinine alkaloids, which reached the minimum level of 5.4 per cent required by the United States Pharmacopeia. In 1940, the farm sent another two samples to the Tonkin Institute of Chemical Analysis in Vietnam, and the results were also quite promising (Table 2). During malaria epidemics in Hekou in early 1940, they distributed some cinchona barks to local malaria patients and collaborated with the local malaria-control station of the YAMC to produce some quinine out of Hekou cinchona bark for malaria treatment. It was reported that many had been cured by taking Hekou-produced bark and quinine, and the staff proudly claimed that 'cinchona bark from our farm is considerably useful'. 78 As a result, Hekou Farm soon enjoyed a boom in the number of visitors, especially those from the central governmental agencies, such as the Army Factory of Medical Materials, the National Drug Research Institute and the National Military Medical School. These visitors had been informed that the Hekou Farm could not promote massive cinchona cultivation due to the limited funds it received from the provincial government, and all agreed that the central government should provide more financial support. One of them even reported this situation back directly to Nationalist China's highest leader at the time, Generalissimo Chiang Kai-shek (1887Kai-shek ( -1975, who then ordered the central government to allocate more funding in early 1940. 79 It was also in 1940 that the National Congress proposed that the central government utilise special medicines produced in southwest China, in particular cinchona, to save more sick soldiers' lives in order to strengthen China's power against the Japanese. 80 As a result of all these efforts, Yunnan's cinchona cultivation programme was incorporated into the central government's national defence project so that they began to receive sufficient funding. In light of this support, Yunnan state-builders thus drafted an ambitious plan, believing that there would be a promising future for their cinchona cultivation project: from 1941 to 1945, the Hekou Farm would cultivate millions of cinchona trees, set up eight new cinchona farms in southwest Yunnan within 5 years and even build a quinine factory in 1951, when there would be adequate grown cinchona trees available. In the first year of 1941, everything was going well: the mature trees in Hekou numbered 3,000, and there were more than 50,000 seedlings. Several new farms had already been established along the Pusi frontier, where cinchona trees had been successfully transplanted from Hekou (Table 3). 81 However, this plan never came to fruition. After the Japanese successfully secured the occupation of French Indochina in late 1941, Hekou became a frontline zone directly facing the Japanese military threat. Its director and main staff retreated to Kunming in the Yunnan interior, and financial support from the central government was also cut off. As a result, the Hekou Farm ceased to exist except in name, and its cinchona cultivation programme was suspended in 1943 and was officially abandoned 2 years later. 82 Only hundred or so of the surviving cinchona trees were then handed over to a local military garrison. 83 The Hekou programme was once again restarted in 1948 by the new Yunnan governor Lu Han , when the Civil War  between the Communists and the Nationalists had already broken out. 84 Soon thereafter with the Nationalist's defeat in 1949, the Hekou farm was taken over by the new Chinese communist regime. In 1953, the Hekou farm was transformed into a new experimental field for rubber plantations to supply 'communist' rubber to the Soviet Union, and its cinchona cultivation programme was entirely abandoned. 85

Conclusion
The past of China's first successful cinchona cultivation programme in Hekou, Yunnan province, has been almost totally forgotten by historians of modern China in general, not to mention Chinese and global historians of medicine. By digging up primary archival sources in Yunnan Provincial Archives, this article reconstructs, for the first time in any language, the origins, developments and the demise of this cinchona cultivation programme between the 1930s and the 1940s, which adds to our understanding of the history of the global spread of cinchona in Nationalist China.
We have gained and are still gaining an increasingly comprehensive understanding of cinchona cultivation in Spanish, British and Dutch empires, 86 but as Ya-wen Ku points out, the East Asian context has been largely overlooked. 87 In her study on Japanese cinchona cultivation in Taiwan, Ku demonstrates the importance of the ignored private sector in the imperial-controlled enterprise. The present article, in elaborating the close relation between cinchona cultivation and the Chinese states (local and national), and the cinchona programme's paradoxical nature of anti-imperialism and internal colonialism, provides another comparison in East Asia, which will also revise and make dialogues with current dominant narratives in the global history of cinchona. The Hekou cinchona cultivation programme was not originally a colonial/imperial project like those in Western or Japanese empires but was initiated by technocrats in Yunnan in the 1930s to eliminate endemic malaria, as part of local developmental state-building. Moreover, their agenda of achieving quinine self-sufficiency against the Dutch turned out to be anti-imperialist in nature. During the Sino-Japanese War , the Hekou programme was integrated into the Chinese central government's national defence project against the Japanese imperialism, as well as to solve broader wartime epidemic crises in southwest China, together with other state-led, nation-wide bioprospecting programmes for medicinal plants. However, its new reorientation to develop the semi-autonomous malaria-ridden Pusi frontier to provide more war sources indicated that the state-led and anti-imperial cinchona programme had ambiguously served as a tool of internal colonialism.
It also contributes to a recent scholarly interest in global histories of medicine, which highlights the role of networks in the global spread of medical knowledge, personnel 85 and objects. 88 By illuminating how Yunnan local state-builders were initially rejected by the Dutch, but later on managed to acquire cinchona seeds from British India, French Indochina and the USA via global botanic cinchona cultivation networks, this article draws attention to the importance of collaborations, as well as conflicts in the global history of medical networks.
Last but not the least, it contributes to ongoing revisions of Nationalist regime  and particularly adds to the growing body of literature on the pivotal period between 1937 and 1945-when China was at war with the Japanese-a crucial decade that has been understudied until recently. 89 They have reminded us that at both the central and local levels, the Nationalist regime was not a total failure or 'an abortive revolution'. 90 As the history of Hekou programme indicates that there was significant progress made even under difficult wartime circumstances and argues that setbacks in the 1940s were primarily due to war and other factors that could not be anticipated or controlled by both the national and local developmental states.