VIII.5

Domestication and the Evolution of Agriculture

Amy Cavanaugh and Cameron R. Currie

OUTLINE

  1. Domestication

  2. Evolution under domestication

  3. Agriculture as a mutualism

  4. Agriculture in ants

  5. Conclusions

Agriculture is an ancient and important factor shaping life on earth. Through the cultivation of food, populations of agriculturalists are able to greatly expand and can even develop a division of labor. This chapter explores the evolution of agriculture, including domestication and selection under domestication, along with the evolutionary events and consequences of farming. It also describes how agricultural associations are perhaps best viewed in the framework of a coevolved mutualism.

GLOSSARY

Artificial Selection. Evolutionary change caused by human breeding in populations of domesticated (or experimental) plants and animals.

Coevolution. Reciprocal evolutionary change between interacting species.

Domestication. Acquisition from the wild of one species by another and breeding it in captivity.

Domestication Syndrome. A suite of traits characteristically found in domesticated species.

Mutualism. An interaction between two species that benefits both.

The domestication of one species by another for food is one of the most significant evolutionary innovations in the history of life on the planet. Indeed, shifting to an agricultural lifestyle, and the concomitant expansion in numbers and range it allows, inexorably alters not only the biology of the species involved but also the ecosystems in which they occur. By establishing a reliable reserve of food, agriculturalists gain an advantage over their hunter-gatherer brethren; the ready source of calories allows the agriculturalist populations to greatly expand and ultimately facilitates the development of a division of labor. Agriculture originated among humans in the Fertile Crescent, but contrary to popular belief, humans were not the world’s first farmers. That distinction belongs to a group of ants in the Amazon Basin. These fungus-farming ants maintain specialized gardens of domesticated fungi that serve as the primary nutrient source for the colony. After the origin of agriculture in ants, but still millions of years before humans appeared, other groups of insects also transitioned to farming. In parallel with fungus-growing ants of the New World, some termites farm fungus for food in the Old World. The most diverse farmers are the Ambrosia beetles, represented by more than 3000 species. In all these cases, the utilization of a farmed food source has enabled these insects to expand into a new ecological niche, leading to their diversification and, in some cases, allowing them to become dominant members of their ecosystems. Other insects engage in more rudimentary forms of farming, and some ants even practice animal husbandry by tending aphids and treehoppers. Besides the insects, a marine snail cultivates fungus, and a species of damselfish farms red algae, and recently it has been suggested that even amoebas practice a rudimentary form of bacterial husbandry.

Agriculture most recently originated in our own species approximately 10,000 years ago and has ultimately resulted in our dominating most of the ecosystems on the planet. Humans cultivate around 100 different plant species, which serve primarily as a reliable and more readily stored source of nutrients. Humans have also domesticated a number of animals, obtaining a variety of benefits, including sources of nutrients (e.g., meat and milk), labor (e.g., plowing fields, transporting of goods, and protecting and herding other domesticated animals), and military advancement (e.g., cavalry). Thus, farming provided a reliable source of calories, allowing an increase in human population size, decrease in birth intervals, and specialization of labor leading to stratified societies, while animal husbandry allowed agricultural societies to expand beyond their borders and ultimately to dominate the nonfarming populations with which they came in contact. Based on these advantages it can be argued, as Jared Diamond does in his Pulitzer Prize–winning book Guns, Germs, and Steel, that agriculture is the single most important force shaping human history.

Just as agriculture has shaped human society and history, it has also had an important role in the development of evolutionary theory. This influence is evident in Darwin’s The Origin, which begins with a thorough discussion of domestication and the evolutionary changes caused by human breeding of domesticated plants and animals—an evolutionary force he termed artificial selection—even before introducing the tenets of natural selection:

It is … of the highest importance to gain a clear insight into the means of modification and coadaptation. At the commencement of my observations it seemed to me probable that a careful study of domesticated animals and of cultivated plants would offer the best chance of making out this obscure problem. Nor have I been disappointed; in this and in all other perplexing cases I have invariably found that our knowledge, imperfect though it be, of variation under domestication, afforded the best and safest clue.

Of the different domesticated species Darwin investigated to “mak[e] out this obscure problem,” the domestic pigeon was the subject of one of his most in depth studies. After thoroughly examining all the breeds of pigeons he could acquire, he determined that more than 20 different characters varied among these breeds. Yet it was believed at the time, and confirmed by his additional studies, that all these diverse breeds had descended from a single wild species, the rock pigeon. Darwin argued that the key factors in creating all this variability among breeds were “man’s power of accumulative selection” and use of the large body of literature, both modern and ancient, in which breeders and horticulturalists described in great detail the ways in which they had modified their animals and plants by selectively mating only those individuals with the desired characteristics.

In this chapter, we discuss evolutionary aspects of domestication and selection under domestication in agriculture by humans. We then argue that a useful way to conceptualize the evolution of agriculture is as a mutualism shaped by coevolution. Expanding on this argument, we end with a discussion on the evolution of agriculture in ants, drawing parallels with humans.

1. DOMESTICATION

Domestication is the practice whereby an organism is acquired from the wild and bred in captivity. The population or species that is domesticated can be referred to as the domesticate. Domesticates undergo genetic changes during the process of cultivation or breeding that make them more useful to the domesticator and ultimately differentiate them from their wild ancestors.

The first domestication of a plant by humans occurred about 10,000 years ago, when people living in the Middle East (parts of modern Iraq, Iran, Turkey, Syria, and Jordan) began to purposefully plant barley, peas, lentils, chickpeas, muskmelon, flax, and two species of wheat. Not long after agriculture had been established in the Middle East, it arose independently in eastern China. There the available wild species differed, and so the first domesticated crops of Southeast Asia included rice, soybeans, adzuki beans, mung beans, hemp, and two species of millet. Populations within the tropical West African and Sahel regions also appear to have independently begun domesticating species including sorghum, millet, rice, cowpeas, yams, bottle gourds, and cotton. Though the dates are uncertain, people in Ethiopia domesticated coffee, and people in New Guinea domesticated sugarcane and bananas. Although populations in the Americas also independently established themselves as farmers, this transition took place later than those in Eurasia and Africa, most likely owing to the inherent differences in the available wild species. Between 9000 and 3000 years ago, humans began domesticating animals including sheep, goats, cattle, pigs, chickens, and horses in Eurasia and northern Africa. Again, populations in the Americas independently domesticated some animal species, such as the llama and the guinea pig, but they were limited in their efforts because most of the available wild species were unsuitable for domestication.

It might seem that almost any wild species could be domesticated, but history has shown that this is not the case. Although humans have domesticated a number of species, they represent an extremely small proportion of the plants and animals that occur in nature. The wild progenitors of the first crop species were already edible, grew quickly and easily, could be stored, and were self-fertilizing. This last trait is crucial in that self-fertilizing plants will directly pass traits on to their offspring largely unchanged. Species that have never been domesticated fail to meet one or more of the preceding criteria. For example, the oak tree, despite producing nutrient-rich acorns, has never been domesticated, for many reasons. First, the oak is an extremely slow-growing tree, taking more than 10 years to grow from an acorn to a fruit-bearing tree. Second, the bitterness of the acorn is under the influence of many genes, which combined with the long generation time, makes it very difficult to select for mutant, sweet acorns. Finally, acorns are a primary food source for another animal, squirrels. By burying large numbers of acorns, squirrels would undermine any human attempt to plant acorns only from oak trees with desirable traits.

Animals that have been successfully domesticated also share many traits. First, most domesticated animals are herbivores. Owing to the successive loss of energy through each trophic level, it takes much less food to support the growth of a herbivore than a carnivore; therefore, raising herbivores is far more efficient. Although we now eat carnivorous fish, we have only recently begun farming them, and whether this leads to their domestication remains to be seen. Second, as with plants, successfully domesticated animals grow quickly. Extremely large mammals, such as elephants, grow too slowly to be candidates for domestication. Third, domesticated species breed readily in captivity. As Darwin noted, this is a particularly rare trait among animals. Fourth, the animal must have a relatively pleasant disposition. While all large animals, and many small ones, are capable of killing humans, most are much more prone to aggression than the species that have been successfully domesticated. Fifth, they must not be prone to panic, particularly panic that results in the animals’ battering themselves to death while trying to escape. This behavioral issue has been a limiting factor in the domestication of many otherwise-suitable herd species, such as gazelles. Finally, many successfully domesticated animals live in herds with well-developed hierarchies and overlapping home ranges; these animals are able to live in proximity to one another and will usually accept a human as the herd leader.

2. EVOLUTION UNDER DOMESTICATION

Although domesticates are species whose wild ancestors possess specific traits suitable for domestication, they are greatly altered by the process of artificial selection imposed by the domesticator. To Darwin, artificial selection was not merely analogous to natural selection but rather represented a clear example of natural selection under a particular set of conditions. The principles are the same, but the environmental conditions in play under artificial selection are those of the human-constructed habitat as opposed to a habitat of nature’s making under natural selection. For either selective force to operate there must be variation in the trait under selection, heritability of that trait, and a tendency for individuals with some version of that trait to reproduce, or be bred, more than others with a different version. As people consciously or unconsciously selected the plants and animals that met human needs and preferentially grew and bred them, they were practicing artificial selection. At the same time, people were creating a novel environment for these plant and animal species, and natural selection further increased the frequency of traits that would lead to success in this constructed environment.

Domestication of plants and animals undoubtedly involved the conscious selection of numerous traits. In plants, early protofarmers likely preferentially collected the largest fruits or seeds to consume and to subsequently plant, and likely selected for taste, choosing the least bitter seeds and sweetest fruits. While many plants were selected for their fruit or seeds, others would have been selected for size or fleshiness of other nutritional parts of the plant (e.g., the roots or leaves), their oil content (e.g., olives and sunflowers), or length of fibers (e.g., flax and hemp). Animals likely were consciously selected on the basis of size, for those raised for meat, or reproductive physiology, for those raised for milk or eggs. Sheep and llamas would have been selected for the retention, rather than shedding, of the wool fibers in their coats, while dogs would have been selected for traits such as size, sense of smell, hunting ability, trainability, and herding ability.

Plants and animals were also subjected to a great deal of unconscious selection. For example, the wild progenitors of cereals and legumes typically drop their seeds as a dispersal mechanism. Mutant plants that did not drop seeds would die out quickly because they would leave no offspring. However, such plants would prove beneficial to humans trying to efficiently gather food, as it is much easier to collect a handful of seeds from the top of a stalk than to pick each individual seed from the ground. Once humans began cultivating plants, selection would have also favored plants with faster germination times. After planting, those plants that sprouted first were more likely to be harvested and replanted, compared with those that delayed germination. Finally, while consciously selecting for traits such as size and taste, humans were also unconsciously selecting for plants capable of self-fertilization. In plants that self-fertilize, as most crops do, favorable mutations are maintained, not diluted by recombination with their neighboring wild progenitors.

Humans attempting to breed the largest or best milk-producing variants of a species would also have inadvertently been selecting for animals with the ability to reproduce in captivity. Domestic animals reach sexual maturity earlier than wild animals and have more frequent reproductive cycles. These traits may have been both consciously and unconsciously selected for by humans—consciously by selectively breeding the animals that reached maturity earliest and breeding them as often as they were receptive to it, and unconsciously by eliminating the nutritional constraints that would have limited their reproduction in the wild.

Together, the forces of artificial and natural selection have led to changes in domesticated plants that have come to be known as the domestication syndrome. These traits include (1) increased size of reproductive organs (e.g., fruits and seeds); (2) increased tendency for mature seeds to remain on the plant rather than dropping to the ground; (3) faster germination as well as synchronized, predictable germination times; (4) changed allocation of biomass (e.g., larger roots, stems, leaves, or buds); and (5) reduced physical and chemical defenses. Domesticated animals also possess a suite of traits that distinguish them from their wild counterparts. Morphologically, domesticated animals typically exhibit greater variation in overall body size as well as in the size of particular body parts (e.g., length of legs in dogs), as compared with their wild ancestors. Additionally, domestic species have different coloration of fur and feathers than their wild relatives, typically an increase in white or spotted coloration. Although such colors make individual animals more visible and therefore more vulnerable to predation, humans could have inadvertently selected for such individuals because they were easy to see and recover if they wandered away.

3. AGRICULTURE AS A MUTUALISM

Agriculture can be thought of as a mutualism—an interaction that benefits both the agriculturalists and the domesticated species. The benefits to humans are obvious, as discussed earlier. But, to some, the benefits of being an “enslaved” plant or animal might not be so clear. However, domesticates do receive numerous benefits, broadly falling into three general categories: (1) protection, (2) increased reproduction, and (3) dispersal. Agriculturalists protect their domesticated crops and animals by significantly reducing interspecific competition, herbivory, and predation. This protection includes growing domesticates in controlled environments and actively weeding, pruning, guarding, and applying chemical treatments. Through the careful planting and cultivation of seeds, farmers increase the probability of seed germination, thus increasing the reproductive rates of domesticated crops. Similarly, domesticated animals have higher reproductive rates, typically owing to shortened interbirth and interlaying intervals. Finally, as agricultural populations spread, they bring their crops and animals with them. By altering the new habitat to be suitable for domesticated species of their homeland, people increase the range of these species. Given the tremendous efforts humans undertake to care for their domesticates and the huge expansion of some plant and animal species following their domestication, Michael Pollan argues in The Botany of Desire that it is worth considering the question, Who is domesticating whom?

Even as humans directed the evolution of the species they domesticated, they created new selection pressures on themselves. The transition to an agricultural lifestyle led to changes in both human behavior and physiology. For example, as with domesticated animals, human agriculturalists have increased reproductive rates compared with those of hunter-gatherers. Most likely owing to the increased reliability of a higher calorie diet, interbirth intervals are much shorter in farming societies than in hunter-gatherer societies. In addition, two enzymes, amylase and lactase, show increased expression in members of agricultural societies compared with hunter-gatherers as well as with chimpanzees, our closest nonhuman relatives. In the case of lactase, an enzyme that digests the sugar found in milk, all mammals produce the enzyme as infants but then stop producing it rapidly after weaning. However, in many human populations, a mutation allows the persistent expression of this enzyme into adulthood. The geographic distribution of this mutation is strongly correlated with pastoralism, particularly the raising of animals for milk production. In a case of parallel evolution, two different mutations have been shown to cause lactase persistence in different populations. Both these mutations occur in the promoter region of the lactase gene. Amylase is an enzyme that breaks down starch. In this case, it appears that populations that switched to the starchier agricultural diet evolved extra copies of the gene that produces salivary amylase. These changes in humans, in response to shifting to an agricultural lifestyle, support the view of agriculture as a mutualism. In fact, they suggest that agriculture represents a mutually beneficial association shaped by coevolution, given that both interactors—the farmer and the domesticate—undergo genetic modification in response to the association.

4. AGRICULTURE IN ANTS

Other than agriculture by humans, the best-studied agricultural association is that of fungus-growing ants. Agriculture in ants is ancient, having originated approximately 45 million years ago. As humans have domesticated many species of plants and animals, fungus-growing ants have domesticated multiple species of fungal crops; there are as many as seven different events of free-living fungi being domesticated. Within this agricultural mutualism the ants and their fungal cultivars have coevolved and diversified. Fungus-growing ants include more than 200 species in 13 genera. Likewise, the cultivated fungi are represented by substantial diversity of strains within specific groups of cultivated lineages. At the pinnacle of evolution of agriculture in fungus-growing ants are the charismatic leaf-cutters, which shape neotropical ecosystems through the sheer mass of leaf material that the ants harvest.

The cultivated fungus, maintained in underground garden chambers in most species, serves as the primary food source for workers, larvae, and the queen. The cultivated fungus produces specialized structures called gongylidia, which are rich in lipids and carbohydrates. The gongylidia appear to represent an optimized nutrient source for the ants, likely evolved under a form of artificial selection. The ants cannot survive without their fungal crops; without them they literally starve. When establishing new colonies, queens ensure the initial presence of the cultivar by bringing a small ball of fungus collected from her parent colony, effectively transferring the fungus from one generation to the next. Recent genomic studies on leaf-cutters have revealed that fungus-growing ants (like humans) have evolved genetically in response to their dependence on agriculture; in particular, they have lost the ability to synthesize an essential amino acid that they likely obtain from the fungus garden.

Leaf-cutter ants have evolved a complex set of behaviors for cultivating the fungus. Like many human-domesticated species, the ants’ fungal crops are unable to survive without the ants. The ants selectively forage for leaf material that promotes the growth of the fungus garden. The garden matrix is thus composed of the fungus and the vegetative substrate that worker ants obtain from outside the nest and then integrate into the fungus garden. Once this leaf material is brought to the colony, the ants lick and chew the material into small pieces. This process breaks down the physical barriers of the leaf that would otherwise prevent the growth of the fungus on the leaf surface. Just as human farmers work manure into the soil, the ants work the leaf pulp into the top layers of the fungus garden. They then bring fungal hyphae from older parts of the garden, plant it onto the surface of the fresh leaf pulp, and continuously add fresh material to the top of the garden.

Besides adding substrate to the garden, the ants also promote the growth of their fungal crop in numerous ways. The ants open and close tunnels to the surface such that they can regulate the temperature and humidity within the growth chambers. There is also evidence that the ants damage the fungus, in a manner akin to pruning, to stimulate increased fungal growth. The fungus produces enzymes that can become disadvantageously concentrated in the garden. When that happens, the ants ingest these enzymes in the areas of high concentration and then defecate them into areas of low concentration, thus creating an equal distribution of the enzymes throughout the garden.

The cultivation of monocultures of clonally propagated crops has led to increased susceptibility to disease. The ants’ fungus garden is host to specialized and potentially virulent agriculture pathogens, microfungi in the genus Escovopsis. Escovopsis—known only from the fungus gardens of these ants, consumes the ants’ fungal cultivar and has coevolved with the ants and their fungal crop. The ants engage in meticulous behaviors to deal with the pathogen. They groom out Escovopsis by pulling pieces of the fungal cultivar through their mouthparts and collecting the invading microbes in their infrabuccal pocket, a cavity and filtering device within the mouthparts of ants. The ants then deposit this material in the refuse chambers. In cases where the garden has become diseased, the ants remove the affected area in a behavior called weeding, which involves ripping out and discarding the infected garden material. Further paralleling human methods for dealing with agriculture pests, the ants employ chemical methods of crop protection. Whereas humans control pests by developing and then spreading chemicals on their crops, the ants form a symbiosis with antibiotic-producing bacteria. These symbionts, Actinobacteria, live on the ants’ cuticle and produce antifungal compounds that inhibit the garden pathogen Escovopsis.

In summary, agriculture in ants, much like human agriculture, has led to their dominant role in many of the ecosystems in which they occur. Further, they share many of the hallmarks of human agriculture, including multiple domestications of wild species, artificial selection of the domesticates, and cultivation including physical and chemical methods for crop protection. Finally, the recent evidence for agriculturally related genetic changes in both the domesticates and the domesticators in human and ant agriculture suggests they represent coevolved mutualisms.

5. CONCLUSIONS

The ability to cultivate and breed plants and animals represents one of the most important developments in human history, allowing rapid and tremendous population expansion. Today domesticated plants and animals consitute an immense proportion of the global caloric intake by humans. Species that have been successfully domesticated share some important characteristics that predispose them to agriculture, and they have undergone significant genetic modification during domestication. Although the changes in domesticated plants and animals have been recognized for millennia, recent work has shown that humans, too, have undergone evolutionary changes in response to agriculture. These genetic changes in humans have occurred in response to farming and consuming specific plants or animals, and they illustrate the coevolutionary nature of agriculture. These general findings have parallels in agriculture by ants, and they show that agriculture and its evolutionary benefits and processes are not unique to humans.

See also chapter VI.7 and chapter VIII.10.

FURTHER READING

Belyaev, D. K. 1979. Destabilizing selection as a factor in domestication. Journal of Heredity 70: 301–308. An experimental study of domestication in the silver fox.

Currie, C. R., J. A. Scott, R. C. Summerbell, and D. Malloch. 1999. Fungus-growing ants use antibiotic-producing bacteria to control garden parasites. Nature 398: 701–704.

Diamond, J. 1999. Guns, Germs, and Steel. New York: W. W. Norton.

Diamond, J. 2002. Evolution, consequences and future of plant and animal domestication. Nature 418: 700–707.

Hölldobler, B., and E. O. Wilson. 2011. The Leafcutter Ants. New York: W. W. Norton.

Pinto-Tomas, A. A., M. A. Anderson, G. Suen, D. M. Stevenson, F.S.T. Chu, W. W. Cleland, P. J. Weimer, and C. R. Currie. 2009. Symbiotic nitrogen fixation in the fungus gardens of leaf-cutter ants. Science 326: 1120–1123. This paper describes another similarity between human crops and ant crops, the need for symbiotic, nitrogen-fixing bacteria.

Pollan, M. 2001. The Botany of Desire. New York: Random House.

Suen, G., C. Teiling, L. Li, C. Holt, E. Abouheif, E. Bornberg-Bauer, P. Bouffard, et al. 2011. The genome sequence of the leaf-cutter ant Atta cephalotes reveals insights into its obligate symbiotic lifestyle. PLoS Genetics 7: e1002007.