The life history of harvester ant colonies

Abstract

A long-term study of a population of desert seed-eating ant colonies of the red harvester ant, Pogonomyrmex barbatus, in New Mexico, USA, shows that a colony can live for 20–30 years—the lifespan of its founding queen. A colony’s collective behaviour shifts in the course of its life history. These changes, generated by social interactions within the colony, adjust the behaviour of the colony as it grows older and larger, in response to its environment and neighbouring colonies. A worker lives only a year and performs different tasks as it ages, in response to interactions with other workers and the local surroundings. A colony’s behaviour changes—becoming more stable and consistent—as the colony grows older, with more ants to participate in social interactions. A neighbourhood of colonies, often of similar age, grows old together. Colonies differ in how they regulate foraging behaviour collectively to manage water loss. These differences influence how foragers of neighbouring colonies partition foraging area. In a harsh but stable environment, the gradual behavioural shifts over a colony’s lifespan allow it to adjust to slow changes in the composition of its neighbourhood and in environmental conditions.

This article is part of the discussion meeting issue ‘Understanding age and society using natural populations’

1. Introduction

An ant colony is a collective individual whose behaviour, resource use and interactions with neighbours all shift in the course of its life history. A long-term study of a population of desert seed-eating ant colonies of the red harvester ant, Pogonomyrmex barbatus, in New Mexico, USA, shows that a colony can live for 20–30 years. Life history changes, generated by social interactions within the colony, adjust the behaviour of the colony as it grows older and larger, in response to its environment and neighbouring colonies. A worker lives only a year and performs different tasks as it ages, in response to interactions with other workers and the local surroundings. A colony’s behaviour changes, becoming more stable and consistent, as the colony grows older, with more ants to participate in social interactions. A neighbourhood of colonies, often of similar age, grows old together. Colonies differ in how they regulate foraging behaviour collectively to manage water loss. These differences influence how foragers of neighbouring colonies partition foraging area. In a harsh but stable environment, the gradual behavioural shifts over a colony’s lifespan allow it to adjust to slow changes in the composition of its neighbourhood and in environmental conditions.

An ant colony operates without central control. The queen merely lays the eggs but does not direct the behaviour of the ants. This insight is at least as old as the writer of Proverbs 6-6: ‘Look to the ant, thou sluggard…who having no chief, overseer, or ruler, provides her meat in the summer and gathers her food in the harvest’. The behaviour of the colony arises from distributed processes using local interactions, mostly olfactory, among individuals [1]. In an ant population, colonies are the reproductive individuals, as colonies mate to produce offspring colonies; this led Wheeler [2] to call a colony a ‘superorganism’. Thus, natural selection can shape heritable variation among colonies in behaviour.

An ant colony’s collective behaviour shifts in the course of its life history. As in many other animals [3], life history changes, generated by social interactions within the colony, alter the colony’s interactions with its environment, including with other colonies. Thus, ants provide a way to learn how the social interactions that generate life history are tuned to the environment. Ant colonies provide an opportunity to see how social interactions shift over the course of life history and how this evolves in response to environmental conditions.

There are more than 14 000 species of ants [4], widespread in an enormous diversity of habitats. All ant species have in common that they live in colonies, consisting of one or more reproductive females along with female workers that do not reproduce.

Surprisingly, little is known about the life history of ant colonies because it is difficult to track colonies over time in natural populations. Queen longevity varies greatly across species (e.g. [5,6]), which means that species differ in the length of time for which data are needed to evaluate colony life history and lifetime reproductive success. For example, a long-term study of a population of red wood ants shows that colonies live for 6–10 years [7], while in the harvester ants discussed here, it is common for colonies to live more than 30 years.

This essay does not attempt to review the literature on the life history of ants, but instead to outline what we have learned about the life history of a single species from a long-term study of a population of colonies of the red harvester ant, Pogonomyrmex barbatus, in New Mexico, USA [8]. In this species, as in others, both ants and colonies change as they grow older. It is an open question whether these changes should be considered as ‘aging’, because there is no evidence of senescence. A colony dies when the queen dies, and all of the workers have died. While we know that old queens are more likely to die than young ones, we do not know exactly why this is. In this harvester ant species, queens continue to reproduce until they die [9].

I will describe three aspects of changes over the life history of a harvester ant colony. First, as a harvester ant worker grows older, it performs different tasks in response to interactions with other workers and the local surroundings. Such transitions from one task to another as a worker grows older—called ‘temporal polyethism’—have been observed in many species of ants [10], as well as in honey bees. As in other ant species, a harvester ant worker first performs tasks inside the nest, beginning with care of the larvae and pupae, and then later, when it is older, works outside the nest in tasks such as foraging.

Second, a colony’s behaviour changes as the colony grows older and larger. An ant’s activity is regulated by interactions with others, and the rate of interaction depends on the local density of ants [11]. As the colony grows in its number of workers, changes in the number of ants available to meet each other can alter the pattern of interactions that each ant experiences and thus their behaviour. In the aggregate, this leads to shifts in colony activity.

Third, the interactions in a neighbourhood of colonies change as they all grow older, so that the age structure and spatial distribution of the neighbourhood influence the survival of all the participating colonies [12–14]. Harvester ant colonies partition foraging area with their neighbours of the same species, and relations among neighbouring colonies develop over years, sometimes decades, of proximity. These relations determine a colony’s food and water supply and thus the colony’s survival and its capacity to produce offspring colonies. Natural selection shapes how individuals respond to social interactions so as to regulate the colony’s foraging activity and its relations with its neighbours.

2. Harvester ant colony life history

Since 1988, I have monitored a population of about 300 harvester ant colonies at a 25 acre site in desert grassland in New Mexico, USA [8]. Harvester ants eat seeds, mostly from annual grasses [15]. A colony of P. barbatus builds a nest with a large mound, often more than a metre across. Every colony at the site is mapped and given an ID number and monitored in an annual census. Each year we find the colonies that were there the year before, note which colonies have died and add the new colonies to the census. From this annual census, we have learned that a colony, consisting of a mated queen and successive cohorts of workers and reproductives, can live for more than 30 years. A colony is founded by a single queen who, in her first summer, joins that year’s population-wide mating aggregation. She uses the sperm from that original mating session to produce all of the workers and reproductives in the colony, decade after decade [16]. Workers live only a year [17]. When the queen dies, the workers gradually die off until the entire colony is gone.

By excavating 12 colonies of known age and counting the ants, we learned how colony size changes over time [18]. The founding queen produces up to several hundred workers in the first year, reaching about 1000 workers by the time the colony is 2 or 3 years old. The colony then grows rapidly in years 3 and 4 and reaches its mature size of 10–12 000 ants when the colony and the queen are five years old [18] (a schematic summary of the results is shown in figure 1). It seems that the colony maintains that size for the rest of the queen’s life. However, because it is not possible to measure colony size without destroying the colony, we cannot evaluate variation among colonies in their rate of growth, and assume that colony size is generally associated with colony age in the manner shown in figure 1, with a sigmoidal growth curve thought to be characteristic of most ant species [19].

Figure 1.

Colony size, in number of workers, with colony age in years. The figure summarizes the results from excavating the nests and counting all the ants in 12 colonies of known age [18].

Because workers live for only a year, all of them must be produced again each year. At about five years of age (the age of both the queen and the whole colony), when the colony reaches its mature size, it begins to reproduce [20]. It sends out winged males and unmated queens—or gynes—to the annual mating aggregation, where the reproductives of all colonies in the population meet. After mating, the males die and newly mated queens fly off at random [9] to found new colonies. This means that offspring colonies do not tend to be neighbours of their mothers.

By matching up mother and daughter pairs of colonies using genetic variation, we were able to produce a life table for the population, showing the number of offspring per colony, when the offspring were founded and how long they lived [9]. These results showed that a colony can continue to reproduce daughter reproductives and sons until the queen dies. Thus, unusually for a terrestrial organism, we found no evidence of reproductive senescence, or any decline in reproductive capacity in older queens. It is not known whether the lack of reproductive senescence is common in ants generally.

3. Changes as individual workers grow older

An ant colony operates without central control. The queen merely lays the eggs but does not direct the behaviour of the ants. The behaviour of the colony arises from distributed processes using local interactions, mostly olfactory, among individuals [1]. Many species of ants have very poor vision and rely mostly on their capacity to distinguish a huge variety of odours [21]. One form of olfactory interaction involves pheromones. Another form of interaction, used by harvester ants as well as many other species, is a brief antennal contact. Ants smell with their antennae, and when one ant touches another, it assesses the odour of the other ant [22]. Like many insects, ants are coated with cuticular hydrocarbons—a layer of a greasy, waxy substance composed of long-chain fatty acids. Harvester ant colonies, like those of many species (e.g. [23,24]), differ in their cuticular hydrocarbon profiles [25] and these differences allow ants to recognize nestmates [26].

Ants within colonies also differ in their cuticular hydrocarbon profiles. As in other ant species, different task groups within a harvester ant colony show characteristic profiles, because an ant’s odour changes in response to its surroundings. The cuticular hydrocarbon profiles of harvester ants are changed in response to conditions outside the nest. As a result, a forager that works outside the nest smells different from a nest maintenance worker, which stays mostly inside the nest [25,27]. Our experiments with glass beads coated with extracts of cuticular hydrocarbons showed that in the course of a brief antennal contact, an ant detects the cuticular hydrocarbon profile of the other ant [22].

An ant uses its recent experience of antennal contacts with particular task groups to decide what to do. This rate of antennal contact can determine whether an ant switches task [28] and also whether it performs a given task actively [29]. In harvester ants, as in other species, individual identity is not important; an ant tracks its rate of contact with other ants of a particular task group, such as foragers, regardless of which individual forager it meets [30].

Individual ants within a colony shift in behaviour as they grow older, beginning with work inside the nest caring for the brood, and moving through a series of tasks until they work outside the nest. An ant’s shifts from one task to another are associated with changes in hormones such as juvenile hormone [31], in biogenic amines including octopamine and dopamine [32] and in gene expression (e.g. [33]).

Task allocation, the process that regulates an ant’s task and whether it is active at that task, is influenced by its rate of interaction with its nestmates [34]. In turn, the rate of interactions depends on the local density of ants in particular task groups [11]. In general, in many ant species, shifts in local density can drive a basic pattern of temporal polyethism, known as ‘foraging for work’ [35,36]. Once the first batch of workers has emerged as adults, and continued throughout the life of the colony, the brood, larvae and pupae are kept together and tended by the youngest workers. As more workers eclose from pupae as adults, the brood chamber becomes increasingly full of workers, so some workers are jostled out of the chamber where they are recruited to other tasks and eventually get pushed out towards the nest entrance to search for food. Harvester ants move from one task to another as they grow older in a way that is consistent with this common pattern of temporal polyethism.

An ant’s task is influenced by its rate of interaction with other ants (e.g. [37]). This interaction rate depends on current conditions. For example, at a time when abundant food is available, a harvester ant forager is likely to meet many foragers and switch to foraging [38]. There is a one-way flow from nest work to foraging [38]. When more foragers are needed—for example, when extra food becomes available—ants from other exterior tasks switch tasks to forage. Experiments with marked individuals showed that the transition to foraging is irreversible; once an ant becomes a forager, it does not revert. An event that elicits more foragers can accelerate the transition from other tasks to foraging, with the result that the colony would have younger foragers than those in one that did not encounter any food windfalls. Similar patterns are well known in honey bees [39].

Social interactions determine not just what task an ant performs, but its current activity in that task. For example, the activity of harvester ant foragers is regulated by positive feedback from ants returning for food [29,40]. A forager goes out on many trips each day to search for seeds [41]. It brings back a seed and puts it down in the entrance chamber, just inside the nest entrance. Whether it leaves the nest on its next trip depends on the rate at which it meets returning foragers with food. This positive feedback is associated with food availability. Each forager searches until it finds a seed [41]. The more food is available, the more rapidly foragers find it and the faster their rate of return with food, stimulating more foragers to leave to search. Foraging ends when foragers stop leaving the nest because it is too hot and dry—at about midday in summer.

4. Aging of a colony

A harvester ant worker lives for only a year [17]. The age distribution of workers within a colony is about the same from year to year because most workers are produced in the summer and die by the fall [42]. However, the total number of ants in the colony changes as it grows older (figure 1). The behaviour of the colony changes with shifts in colony growth and the need for food to support it. Younger, smaller colonies act differently from older, larger ones in several ways. First, older, larger colonies of harvester ants are more stable and consistent than younger, smaller ones in their response to changes in conditions such as food availability and the amount of refuse to be cleared from the nest entrance [1,43]. This may be a consequence of colony size. The larger the colony, the more ants are available to meet other ants. We do not know exactly how the dynamics of interactions generates stability, but it is plausible that more frequent interactions may lead to more consistent outcomes.

Colonies also change how they interact with neighbouring conspecific colonies as they grow older and larger. Colonies compete for foraging area with their neighbours of the same species. When foraging trails of neighbouring colonies overlap, fighting is rare [44], but the food that one colony takes is lost to the neighbour. Colony interactions with neighbouring conspecific colonies shift with colony life history [18,44]. A colony’s foraging area does not scale linearly with colony age [20,45]; the foraging trails of a 3- to 4-year-old colony are about as long as those of a mature, older and larger one, reaching up to 15−20 m from the nest. However, interactions with neighbouring conspecific colonies change as the colony grows older and larger.

In an experiment investigating how a colony’s age influences its interactions with its neighbours, nest mounds were enclosed by surrounding them with a barrier that prevented foraging. The foragers from neighbouring colonies soon began to search in the former foraging area of the enclosed colony [18]. Then the enclosure was removed. How a colony reacted once its foragers were released, and met the encroaching foragers of their neighbours, depended on its age. Foragers of very young colonies, 1−2 years old, were likely to retreat when they encountered foragers from a neighbouring colony. By contrast, the foragers of a 3- to 4-year-old colony—during the period of rapid growth before the colony reaches its mature size and begins to reproduce (figure 1)—were likely to engage in repeated confrontations with the foragers from a neighbouring colony. Colonies that had reached reproductive maturity, i.e. aged 5 years or older, were likely to avoid trails that overlapped those of a neighbour.

Social interactions within a colony may influence how a colony’s growth shapes its relationship with its neighbours. Most of an ant colony’s food goes to feed the larvae, and the behaviour of foragers is influenced by the colony’s need for food in many ant species [46]. Life history changes in a harvester ant colony’s interactions with its neighbours may be related to changes in the need for food. Since workers live for only a year, the foragers of an older colony are not older than those of a younger one. But the ratio between the number of larvae requiring food and the number of adult workers available to obtain food for them differs greatly over the course of the colony’s life. A 3- to 4-year-old colony has about 6000 ants, but it must produce about 8000 workers by the following year and thus has a higher ratio of larvae to foragers than an older, larger colony that has 10 000 ants available to produce another 10 000 the following year (figure 1). This may be why, in a 3- to 4-year-old colony, foragers may experience more pressure to obtain food and thus they may be more likely to continue foraging despite interactions with ants of neighbouring colonies.

5. Colony neighbourhoods

Because colonies partition foraging area with their neighbours, social interactions among neighbours are crucial for colony survival. A young colony’s chances of survival depend on the age and proximity of its neighbours [8,44]. After the annual mating flight, newly mated queens disperse at random to found new nests [9] and there is no evidence that queens assess the neighbourhood before choosing a nest. Some land on the nests of other colonies and are quickly killed. While in some years many hundreds of new nests are founded at the site, only about 5% survive.

A new colony is unlikely to survive into the following year if its founding queen chose a nest site close to a large old colony [8,44]. On average, a new colony must be at least about 8 m from an existing one to survive, and the younger and smaller the neighbours, the greater its chance of survival.

Changing climatic conditions are influencing how the spatial configuration of a neighbourhood and the ages of its member colonies determine their survival. The configuration of the neighbourhood has an impact year after year, as the colonies all grow older. Beginning in the second decade of the long-term study, rainfall in the southwestern US began to decline. We examined the effects of the spatial configuration of the local neighbourhood of conspecifics on colony survival and recruitment of new colonies [8]. In the earlier years of the study, when rainfall was high, colony numbers increased and then began to decline after about 1997−1999, apparently owing to crowding. As rainfall decreased, beginning in about 2001−2003, both the recruitment of new colonies and colony survival declined, leading to a trend towards earlier colony death. The decrease in rainfall apparently led to a decrease in food availability. This led to a greater impact of crowding, such that more foraging area was needed to sustain a colony. Although overall the number of colonies declined once the drought began, so colonies have become less crowded, the scarcity of food in a crowded neighbourhood has become more dangerous, especially for young, small colonies.The configuration of the neighbourhood has an impact year after year, as the colonies all grow older.

6. Variation among colonies

Colonies regulate foraging activity to manage water loss. Ants lose water to evaporation while out foraging in the sun. Their water comes from metabolizing the fats in the seeds they eat. So a colony must spend water, in foraging time, to obtain both water and food.

All colonies face this trade-off between water loss and food supply, but colonies differ in how they do this. When it is overcast and humid, or on the day after rain, all colonies tend to react in the same way by ramping up foraging activity [47]. It costs less water to obtain food when it is humid because an ant can walk around searching for seeds without losing much water. In addition, after rain there is more food available because the upper layer of the soil is washed away, exposing more seeds for the ants to find. So on the day after rain, during the summer monsoon season when plants are flowering and setting seed, it is a foraging extravaganza, with ants pouring out of all the nests and streaming back in with seeds.

But when conditions are dry, there is more at stake. If there has not been much rain then there might not be much food, and the drier the air, the more water will be lost by ants outside searching. Is it worth it to go out on the next trip? Colonies differ more in foraging activity on dry days than on humid days [47]. In some colonies, foragers make fewer trips on dry days; apparently, when humidity is low, it requires more food availability and more interactions with returning foragers to stimulate foragers in such colonies to leave the nest. Reducing foraging activity in dry conditions sacrifices food intake but conserves water while the colony relies on stored food.

Many colonies are consistent, from year after year, in how they manage water loss. We made longitudinal observations of foraging activity in 95 mature colonies over 5 years, between 2016 and 2021. Within a year, some colonies tended to reduce foraging on dry days while others did not. These differences among colonies in collective behavioural plasticity—in how they regulate foraging activity in response to day-to-day changes in humidity—persisted from year to year [47]. Approximately 40% of colonies consistently reduced foraging activity, year after year, on days with low humidity; approximately 20% of colonies never did this, foraging as much or more on dry days as on humid days. The rest differed in behaviour from year to year.

Because workers live for only one year, the persistent colony behaviour from year to year suggests that how a colony regulates foraging activity arises from physiological factors that foragers inherit from their mother queen and her mates. How a colony regulates foraging is associated with (at least) two physiological factors: the effectiveness of a colony’s cuticular hydrocarbons in preventing water loss and the neurophysiology of dopamine. Ants in colonies that reduce foraging in dry conditions lose water faster than those from the colonies that do not [48]. We are currently investigating how this is related to differences in the chemistry of a colony’s characteristic cuticular hydrocarbon profile [49]. A second source of variation among colonies in the regulation of foraging activity is related to dopamine, which mediates a forager’s decision whether to leave the nest on its next trip [48,50].

Thus, differences among colonies in physiological characteristics lead all foragers within a colony to make similar decisions about foraging in relation to humidity [51] in response to social interactions. In the aggregate, forager decisions about whether to leave the nest on the next trip determine both the colony’s food supply and its water loss. We are currently investigating the heritability, from parent to offspring colony, of the physiological factors that influence forager decisions.

Colony variation in the regulation of foraging shapes the ongoing influence of the configuration of colony neighbourhoods as colonies age. Because young colonies are less likely to survive near old, larger ones, neighbourhoods are likely to be comprised of similarly aged colonies. Like trees in a forest, the old trees shade the younger ones, and so a neighbourhood of young trees may all grow to reach the sunlight together. In the same way, a group of young colonies can each claim a foraging area as they grow, leading to a neighbourhood of colonies of similar age.

The long-standing relations among neighbouring colonies that continue over decades determine whether a colony can survive. In a neighbourhood of colonies that reduce foraging when dry, colonies can grow old together while conserving water; all the colonies will do better. However, if a colony that reduces its foraging during dry conditions is surrounded by others that do not, it risks losing foraging area to its neighbours while it is conserving water. Even if all of the neighbouring colonies are of a similar age, the one that reduces its foraging may be at a disadvantage. By contrast, if all the colonies in the neighbourhood spread out to search for food regardless of the weather, on dry days as well as humid ones, the foraging area of each one will be smaller because foragers go only as far as they can before they meet the neighbours. If a colony’s foraging area is too small, when it extends its foraging activity on a humid day it faces more encroachment from its neighbours, which are likely to be active on that day as well.

7. Natural selection on life history changes in collective behaviour

How individual foragers respond to the social interactions that regulate foraging influences colony lifetime reproductive success, in offspring colonies, as the colony grows older. In a study conducted in 2010, early in the current drought, selection favoured colonies that reduced foraging in dry conditions. These colonies had higher lifetime reproductive success [52]. However, since 2010, when reducing foraging during dry conditions was favourable for colony reproductive success, the drought in the southwestern US has deepened. The food supply may become so low that colonies cannot afford to save water, especially if they grow old in a neighbourhood of colonies that do not reduce foraging to conserve water and thus claim more foraging area.

As the colony grows older, its collective behaviour changes, because its behaviour is mediated by social interactions among individuals, and the number of individuals increases. A colony grows slowly, over 5 years, and then persists for another 25, from year to year adjusting to the foraging behaviour of its neighbours and the effects of sparse rainfall on its food supply.

The life history of harvester ant colonies, like the rest of its collective behaviour, reflects the harsh and relatively stable environment in which the species is evolving [53]. Further studies of the life histories of other ant species will probably reflect similar responses to ecological conditions [54]. For example, it seems likely that species in rapidly changing environments may reproduce more frequently than once a year, as in many tropical species [55], and possibly have shorter lifespans, in response to frequent fluctuations in resource availability.

To learn more about how natural selection shapes collective behaviour over colony life history, studies are needed that track colonies over their lifetimes and examine colony reproductive success. Continuing the many ongoing excellent studies of life history in a diversity of ant species (e.g. [56–59]), can show whether there are ecological patterns in the many different ways that ant colonies of different species change their collective behaviour over the course of a colony’s life.

Ethics

This work did not require ethical approval from a human subject or animal welfare committee.

Data accessibility

This article has no additional data.

Declaration of AI use

I have not used AI-assisted technologies in creating this article.

Authors’ contributions

D.M.G.: conceptualization, investigation, methodology, and writing—original draft.

Conflict of interest declaration

I declare I have no competing interests.

Funding

No funding has been received for this article.