A High in the Andes of South America, in the region around Lake Titicaca on what is now the border between Peru and Bolivia, people began cultivating the potato at least 7,000 years ago. It was an unlikely candidate for domestication. Most wild potatoes are small and knobbly, and they are laced with glycoalkaloids, bitter compounds that can cause serious illness if consumed in quantity. Archaeologists believe that early farmers gradually favoured plants whose tubers were less toxic, and some communities also discovered that a particular kind of clay, eaten alongside the tubers, binds to the poisons and renders them harmless — a practice still observed in parts of the highlands. Over the following millennia, Andean cultivators developed an extraordinary range of types adapted to different altitudes, soils and microclimates; several thousand native varieties survive in the region today.

B Andean farmers also solved a problem that has troubled every potato grower since: fresh tubers rot within months. Their answer was chuño, a freeze-dried potato produced by exploiting the sharp contrasts of the highland climate. The harvest is spread out on open ground to freeze during bitterly cold nights and then thawed by the daytime sun; after several days, the softened potatoes are subjected to trampling, which squeezes out the remaining moisture. The result is a light, hard product that can be kept for years without spoiling. Under the Inca, who ruled much of the Andes in the fifteenth and early sixteenth centuries, chuño was kept in vast state storehouses and used to feed labourers and soldiers.

C Europeans first encountered the potato in the 1530s, when Spanish expeditions pushed into the northern Andes. Within a few decades it had crossed the Atlantic: it was being grown in the Canary Islands by the 1560s, and records show that a hospital in Seville was buying potatoes in 1573. From Spain the plant spread slowly to Italy, the Low Countries and the British Isles. For much of the next two centuries, however, it was treated more as a botanical curiosity than as a food. It appeared in herbals and in the gardens of wealthy collectors, but ordinary farmers were reluctant to plant it.

D The reasons for this reluctance were partly practical and partly cultural. Varieties brought from the equatorial Andes were adapted to days of roughly equal length, and in northern Europe they often produced tubers only late in the autumn, if at all; it took many generations of selection before plants suited to long summer days became common. In addition, the potato belongs to the nightshade family, several of whose members are notoriously poisonous, and its own leaves are indeed toxic. Rumours circulated that eating it caused leprosy, and some critics pointed out that it was mentioned nowhere in the Bible. Acceptance came largely through official pressure. In 1756, Frederick the Great of Prussia ordered his subjects to grow the crop, and in France the pharmacist Antoine-Augustin Parmentier, who had survived on potatoes as a prisoner of war in Prussia, campaigned tirelessly on its behalf. According to a popular and possibly embellished story, he had a potato field guarded by soldiers during the day, reasoning that anything so carefully protected must be valuable, and then withdrew the guards at night so that local people could steal the crop.

E Once established, the potato transformed the diet of much of northern Europe. An acre planted with potatoes can yield considerably more food energy than the same area under wheat or rye, and the crop is ready to harvest within three or four months. Potatoes also supply vitamin C, which grain lacks, and because the edible part grows underground, a field is less easily destroyed by passing armies than a field of ripening corn. In 2011 the economists Nathan Nunn and Nancy Qian estimated that the introduction of the potato accounted for about a quarter of the growth in population and urbanisation in the Old World between 1700 and 1900.

F Yet dependence on a single crop carried grave risks, as Ireland discovered. By the early nineteenth century, roughly a third of the Irish population relied on the potato for almost all of their food, and an adult labourer might eat several kilograms a day. Much of the crop consisted of one variety, the Lumper, which yielded heavily on poor land but offered little resistance to disease. In 1845 a pathogen called Phytophthora infestans, which causes late blight, reached Ireland, turning tubers into black, rotting pulp both in the fields and in storage. The crop failed repeatedly over the following years. By the early 1850s around a million people had died of starvation and related diseases, and at least as many again had emigrated; the population fell by roughly a quarter. Historians stress that the scale of the disaster owed as much to political decisions as to the blight itself: food continued to be exported from Ireland throughout the famine.

G Today the potato is the world's third most important food crop for direct human consumption, after rice and wheat, and global production exceeds 350 million tonnes a year. China is the largest producer, followed by India, and the crop is spreading fastest in the developing world. The lesson of the Irish famine has not been forgotten. At the International Potato Center in Lima, Peru, scientists maintain a gene bank holding thousands of samples of native and wild potatoes, which breeders can use to develop varieties able to withstand new diseases and a changing climate.

A Walk through almost any forest and the most important living system may be the one you cannot see. Threaded through the soil are the hyphae of mycorrhizal fungi — filaments far finer than a human hair — which colonise the roots of plants and extend outwards into the surrounding earth. The partnership is ancient: fossils from the Rhynie chert in Scotland, roughly 400 million years old, preserve fungal structures inside the tissues of some of the earliest land plants, and many biologists believe that plants could not have moved out of the water without such allies. Today, an estimated 80 to 90 per cent of plant species form mycorrhizal associations.

B The arrangement is fundamentally a trade. Because hyphae are so slender, they can penetrate pores in the soil that are closed to even the finest roots, and they explore a far greater volume of earth for the same investment of material. The fungi deliver water and mineral nutrients, especially phosphorus and nitrogen, to their hosts; in return, the plants supply sugars produced by photosynthesis. The price is not trivial. Estimates vary widely, but a plant may pass a substantial fraction of the carbon it fixes — sometimes put at a fifth or more — to its fungal partners, which suggests that the benefit it receives is considerable.

C Because a single fungus can colonise the roots of several plants at once, and a single plant may host many fungi, it has long been suspected that neighbouring plants might be linked into what researchers call a common mycorrhizal network. The idea attracted wide attention in 1997, when the forest ecologist Suzanne Simard and her colleagues published a study in the journal Nature. Working in British Columbia, they supplied paper birch and Douglas fir seedlings with carbon labelled with different isotopes, and detected carbon moving between the two species, with the shaded fir apparently receiving a net gain. The journal's cover described the finding as the 'wood wide web', a phrase that has stuck.

D Over the following two decades, the network became one of the most compelling stories in popular science. Later work in Simard's laboratory used genetic markers to map the fungi linking Douglas firs in one stand; the most connected tree was linked to dozens of others, prompting the image of 'hub' or 'mother' trees that nurture their offspring. Books such as Peter Wohlleben's bestseller The Hidden Life of Trees and Richard Powers's Pulitzer Prize-winning novel The Overstory carried these ideas to millions of readers, often presenting forests as cooperative communities in which trees share food, recognise their relatives and send warnings of insect attack through the soil. Simard's own memoir, Finding the Mother Tree (2021), later reached a similarly wide readership.

E Not all researchers were comfortable with this picture, and in 2023 their doubts were set out in detail. Justine Karst of the University of Alberta, together with Melanie Jones and Jason Hoeksema, reviewed the field studies underlying three widespread claims: that such networks are common in forests, that resources moving through them improve the performance of seedlings, and that mature trees preferentially send resources or warning signals to their kin. They concluded that none of the three was well supported by the available evidence. Many experiments, they noted, could not rule out alternative routes: carbon or chemical signals might travel through the soil itself, or via root contact, rather than through fungal threads. Results were also inconsistent; in some trials, seedlings connected to larger trees grew worse rather than better.

F Perhaps the most striking part of the critique concerned how science is reported. Karst and her co-authors found that unsupported or overinterpreted statements about the networks had appeared in a growing share of research papers, and that studies with positive results were cited more often than those without, a pattern they termed positive citation bias. In their view, an attractive narrative had run ahead of the data, and repeated citation had gradually converted tentative findings into apparent fact. Simard and others have responded that the critics demand an unrealistic standard of proof from field research, where perfectly controlled experiments are rarely possible, and that a large body of evidence still points towards meaningful exchange.

G Underlying the disagreement is a practical difficulty. Proving that two trees are connected requires showing that the same individual fungus, identified genetically, is present on both sets of roots, and then demonstrating that material actually moves through that fungus rather than around it. Only a handful of forests have been examined in this way. Such work is slow and costly, not least because the roots of a single tree may host dozens of fungal species, many of which can be reliably told apart only by laboratory analysis. Researchers on both sides agree that mycorrhizal fungi are vital to plant health and to the storage of carbon in soil, and initiatives such as the Society for the Protection of Underground Networks, founded in 2021, are now attempting to map fungal communities across the globe. What remains uncertain is not whether the fungi matter, but whether forests behave like the caring societies of popular imagination.

A In 1945, as the Second World War drew to a close, Vannevar Bush, who had directed the United States' wartime research effort, delivered a report to President Truman entitled Science, the Endless Frontier. Its central argument was that basic research — inquiry pursued without any particular application in mind — was the 'pacemaker of technological progress', and that because private industry would never pay for enough of it, the federal government should. The report helped to bring about the creation of the National Science Foundation in 1950, and its assumptions have shaped science policy across much of the world ever since. Nearly eight decades later, however, the question it appeared to settle is being asked again: should the state be the principal paymaster of science, or would research be better served by a more varied set of patrons?

B The economic case for public funding was given its classic form in 1962 by Kenneth Arrow. Knowledge, he observed, is unlike most goods: once produced, it can be used by any number of people without being used up, and it is difficult for the discoverer to prevent others from benefiting. The predictable result is that markets invest less in research than would be best for society as a whole, and the shortfall is greatest for precisely the kind of open-ended work whose eventual uses cannot be foreseen. On this reasoning, government support is not a subsidy to scientists but the correction of a market failure.

C Not everyone accepts the argument. The British biochemist Terence Kealey has argued, most fully in his 1996 book The Economic Laws of Scientific Research, that the market failure is largely imaginary. Firms, he contends, already fund a good deal of basic science, partly because they need in-house expertise to understand and exploit the discoveries of others; and when governments step in, they tend to displace private spending rather than add to it. Critics respond that his historical examples, drawn largely from the nineteenth century, are hard to interpret, since the scale and cost of modern research bear little resemblance to those of the Victorian laboratory.

D A quite different defence of public funding comes from the economist Mariana Mazzucato. In The Entrepreneurial State (2013), she argues that governments have not merely filled gaps left by business but have actively created new markets by taking on risks that private investors would not. Much of the technology inside a modern smartphone — the internet, satellite positioning, touchscreen displays, voice recognition — originated, she notes, in publicly funded research. Her complaint is less that the state does too much than that it is rarely rewarded: the public bears the early risks while the profits flow largely to private companies. She has therefore proposed mechanisms, such as retaining equity stakes, through which taxpayers might share in the returns.

E Meanwhile, the landscape of funding has become more mixed than any of these theories implies. In most wealthy countries, businesses now pay for the majority of all research and development, although much of that is applied development rather than basic science. Philanthropic bodies such as the Wellcome Trust and the Gates Foundation spend billions annually. Even the most celebrated contest between public and private science — the race to sequence the human genome, in which an international publicly funded consortium was challenged in the late 1990s by the company Celera Genomics — ended in a joint announcement in 2000, and arguably showed that each side spurred the other on.

F Public funding has weaknesses of its own, and its defenders do it no favours by ignoring them. Most government money is distributed through competitive grants judged by panels of established scientists, a system that is inclined to reward safe, incremental proposals over risky ones. The career of Katalin Karikó, whose work on modified messenger RNA was repeatedly turned down for funding in the 1990s, before the same research underpinned COVID-19 vaccines and earned her a share of the 2023 Nobel Prize, has become a cautionary tale. The process is also extravagant in its demands: one study of a single annual round of Australia's main medical research grants estimated that applicants had collectively spent more than five centuries of working time preparing proposals, most of which were unsuccessful.

G Some funders are experimenting with remedies. Since 2015, the Health Research Council of New Zealand has allocated some of its grants by lottery among applications judged to meet a basic standard of quality, and the Swiss National Science Foundation has used random selection to decide between proposals of similar merit. Others favour funding promising people rather than particular projects, giving researchers long-term support and the freedom to change direction.

H My own view is that the question is wrongly framed as a choice. Private and charitable money is welcome, but it flows towards problems that are fashionable, profitable or close to a donor's heart, and it can be withdrawn on a whim; only governments can sustain the patient, unglamorous inquiry whose value may not be apparent for decades. The answer to the defects of public funding is therefore to reform it, not to retreat from it — to take more risks, to reward boldness and to spend less of scientists' lives on paperwork. Bush was right that basic research needs a public patron. Where he was naive was in assuming that generosity alone would guarantee good science.