Coffee Is Not Forever. Stuart McCook
farmers tended to farm coffee under shade and at lower altitudes, sometimes as low as sea level. “In these cases,” wrote Hull, “the plants will invariably be found growing under the shade of the jack, cocoa-nut, or other suitable trees, without which protection all chance of their thriving permanently would be out of the question.” These farms were also “limited in extent, and are generally richly manured and often well watered during the dry season.”3 According to some accounts, they also cultivated coffee in small patches around their villages. James Webb persuasively argues that during the 1820s and 1830s, some Sinhalese farmers adopted more intensive production techniques, supported by a series of policy changes by the colonial government to encourage local production. This intensification involved clearing highland chena lands and forests to take advantage of their rich soils. Webb estimates that they must have cleared some 60,000 acres of forest for coffee production, sometimes bringing them into conflict with British coffee planters.4
Although the earliest British estates were established in the 1820s, the estate boom did not began in earnest until the 1840s. Spurred by rising coffee prices in Europe and North America, Europeans (mostly Britons) started aggressively clearing the steep slopes of Ceylon’s highland forests. In 1840, the colonial government decreed that Ceylon’s highland forests belonged to the Crown. Over the next several decades, much of this land was sold to settlers for coffee production. Most of the colonists had little previous experience with farming coffee, or indeed, farming of any kind. Like estate coffee producers in other parts the world, they sought to maximize productivity and profitability. “It is generally admitted,” observed the planter William Sabonadière, “that nothing equals virgin forest land for the cultivation of coffee.” The trees were cleared so that coffee could be cultivated on the rich forest soils. At first, farmers often obtained arabica seeds and seedlings from neighboring Sinhalese farms. The estates were usually planted without shade, since shade reduced yields. By the late 1880s, 600,000 acres of forest had been cleared for coffee. “The lovely sloping forests are going,” wrote the novelist Anthony Trollope on a visit to Ceylon, “and the very regular but ugly coffee plantations are taking their place.” Between 1849 and 1868, annual coffee exports tripled, from roughly 330,000 hundredweight to 1,000,000 hundredweight (16,700 metric tons to 50,800 metric tons).5
The “most fruitful coffee districts,” according to Edmund Hull, were in the highlands between 2,500 and 3,500 feet (roughly 750–1,050 meters), although estates could be found at altitudes between 50 and 1,500 meters. Like coffee planters everywhere in the early nineteenth century, farmers on Ceylon (both Sinhalese and European) pushed arabica coffee to its ecological limits. They managed the plant as best they could by manipulating the conditions under which it was cultivated. In these respects, Ceylon was just as vulnerable to the rust as other coffee zones around the world. In another respect, however, it was even more vulnerable. Ceylon’s climate was unusually wet and windy. The summer monsoon (May–September) and the winter monsoon (December–February) showered the island with rain and exposed it to winds that could exceed 100 kilometers per hour. Rain also fell regularly during the intermonsoonal period. Ceylon’s wet climate, then, turned its coffee farms into a vast incubator for the coffee rust. And the strong and regular winds ensured that when the rust appeared, it would spread rapidly through the island and beyond.6
The Search for Origins
The early history of the rust in Ceylon remains murky. Some observers contended that the rust fungus was native to Ceylon, arguing that the epidemic had been triggered by the introduction of C. arabica. The rust’s supposed wild host was a plant then known as Coffea travancorensis (now classified as Psilanthus travancorensis), a plant closely related to coffee and indigenous to Ceylon and Southern India. Thwaites, who had been studying Ceylon’s fungi for almost a decade, argued that this was unlikely because he only found the rust on C. travancorensis after the epidemic had already broken out. It was more likely that C. travancorensis was infected from C. arabica, not the other way around. The naturalist John Nietner, along with several others, suggested that the rust had likely been present on coffee farms for several years before the epidemic broke out. Thwaites countered that if the rust had been present, “it is somewhat remarkable that the somewhat conspicuous orange-coloured spores on the underside of the leaves did not attract attention; and it is equally remarkable that the disease should so suddenly have assumed so very malignant a character.”7 Thwaites could not have known this, but as the rust later spread around the world, it often did pass unnoticed for several years before attracting attention—even when people were specifically looking for it.
This last objection is the most compelling one against an early introduction of the rust to Ceylon: if the fungus had been present on the island for any significant length of time, why didn’t the epidemic break out sooner? All the ecological conditions for a large-scale rust epidemic had existed since at least the coffee craze of the 1840s. One possibility is that the fungus had been present in Ceylon for some time, but the arabica plants cultivated in Ceylon were resistant to the strain of H. vastatrix that was initially introduced. After a time, a new, more virulent strain of the fungus evolved that could overcome that resistance. Later in the story, we shall see examples of this pattern. But as far as we know, most cultivated arabica is susceptible (to a greater or lesser degree) to all strains of H. vastatrix, and many, if not most, plants would surely have shown at least some lesions. It is hard to understand how it could have escaped notice altogether before 1869.
The location and pattern of the early rust outbreak in Ceylon strongly suggest that the rust was introduced. It began at a single point—Madulsima—in the interior of the island and spread outward from there. This pattern is characteristic of what plant pathologists now call a “focal epidemic,” which typically begins with a low level of inoculum (in this case, spores) at a well-defined location. The fungus then spreads outward from the focus in waves, like ripples in a quiet pond after a stone is dropped in.8 Had the fungus already been widespread in Ceylon’s forests, it is highly unlikely that the disease would have appeared at such a well-defined location; the epidemic would have been generalized across Ceylon’s coffee farms from the very beginning. But if we accept that the fungus was introduced, then we need to ask how.
In the mid-nineteenth century, pathogens of all kinds were traveling farther and faster than ever before. The historian David Arnold has aptly described the Indian Ocean basin in those years as a “disease zone” in which pathogens of all kinds followed the ebbs and flows of empire and found new host populations on which to survive and reproduce.9 Ceylon was tightly linked into a global network of steamships that regularly and swiftly moved goods and people between Asia, Africa, the Pacific, and beyond. New innovations like the Wardian case—essentially a portable greenhouse—made it possible for people to ship live plants anywhere in the world.10 This increased the risk of accidentally moving diseases and pests that fed on those plants. Newly empowered public institutions, such as the Royal Botanic Gardens at Kew, helped broker the movement of plants and seeds across the global tropics. Private nurseries, such as William Bull and Sons in London and the Horticole Coloniale in Brussels, also supported a global trade in tropical seeds and plants. The geographic and economic barriers that had for several centuries kept the coffee rust contained in Africa had begun to erode.
In the 1980s, the biologist Gordon Wrigley speculated that the spore might have been brought to Ceylon by the Napier expedition, a military expedition that sent troops from India to Ethiopia early in 1868. While in Ethiopia, the expedition passed through a number of minor coffee-growing areas where the rust may have been present. A number of people on the expedition, including Napier himself, had close connections with Ceylon. Wrigley speculates that they “might have returned with some live or pressed coffee plant material carrying viable spores.”11 This is one possible explanation. But the fungus need not have come from Ethiopia; it was also present on wild coffees in the Great Lakes region of East Africa and in the upper reaches of the Congo. Trade routes for slaves and ivory passed through these regions. These routes linked the East African interior to maritime trade networks spanning the Indian Ocean and beyond. They were heavily traveled by many people, including African slaves, Arab traders, and European missionaries and explorers, among others. Perhaps one of these travelers brushed up against some infected coffee plants and inadvertently picked up some spores