Army Ants: A Study Of Social Behavior (1966)

of nomadic marauding tropical American ants whose patterns of behavior have been evolving for more than 50 million years. They represent a highly advanced form of social behavior, knowledge of which contributes to our general theories of animal behavior. How can we understand the organization of colony life in these ants? The scientist's way is to make observations in the natural environment, to repeat an observation many times, and to conduct experimental tests in the field laboratory under controlled conditions over a long period of time. Following an army ant column in the dim light and tangled growth of the forest is not an easy course, but it is the only way to find the home site or bivowak of the colony. A colony consists of a queen and many thousands of workers. This golden yellow ant is the surface adapted species esetan hammatum. By clustering together, the workers form the walls and inner structure of the nest with their own bodies. No other materials are used except the natural supports to which they cling. Nearly all the ants hang head downward in the structure. Opposed double hooks at the end of each leg enable the ants to hold on to different surfaces and to interlock their bodies. The bivwack or temporary nest is the place where the study of these insect colonies begins. From each bivvoac, the colony carries out its great raid of the day and immigrates at night. How can activities so complexely organized and highly adaptive as these be accomplished by creatures as simple as army ants are in their individual behavior? We will start our observations with the beginning of activities in early morning. When the first light reaches the bivwac wall, it stimulates and excites the ants. Once aroused, the workers begin to leave the cluster for the ground. This army ant colony is the species known as esetan burchchelli. It too is adapted to live above ground. This bewac is about 1 meter in width. In this observation, blowing on the bivwac wall excites and agitates the ants. Wind or rain would have a similar effect, causing the bivwac to shift a little. Each morning, workers numbering some hundreds of thousands, as in this species, swarm from the bivwak to pillage for food on the forest floor. As their numbers increase on the ground, the ants are pushed ahead by the pressure of those behind. The sounds you hear are those of ant birds. They eat insects stirred up by the foraging ants. Their calls are a useful clue for locating an ant swarm. During the great exodus from the bivwac, at the beginning of the raid, the ants swarm out in all directions from the nest. Gradually, this movement turns toward a dominant sector. Ants coming in from other areas move in columns and connect with the developing swarm or the bivwac. As the ants move farther from the bivwac, the pattern of the raid takes a roughly triangular or fan-like form with only a single broad column connecting to the bivwac. At the height of the raid, the ants in the swarm front may cover an area more than 15 m wide and 1 to 2 m deep. They stream along the ground, probing under every leaf and in every cranny for practically any living creatures. Unless very agile, few softbodied insects escape the fastmoving ants. Few obstacles in the terrain stand in the way of their attack. They advance swiftly across the forest tangle, even climbing to the topmost branches of trees, capturing their prey. Whenever an insect such as this Katy did is overrun, it is torn to pieces so that it can be easily carried. By pulling together from all sides, the ants dismember their booty. They attack anything that moves. Under the weight of ants, most efforts to escape are feutal. This large and predatory ponerine ant will be fought, stung, and then taken apart limb by limb. The buzzing of flies of various species is another characteristic sound accompanying a swarm raid. By midm morning, the booty laden ants crowd the principal trail as they move toward the bivwac. Parts of insects and other arthropods such as eggs, larvas, and pupas of wasps and ants are hauled along in the procession. How do the ants find their way with all the confusion that seems to exist in these masses of scurrying bodies? There are no leaders yet. At the height of the raid, the ants may be out 90 m or more from the bivwack. Here's a simple test that may give us a clue. Why do the ants stop short when the leaf is removed? Why do they continue when it is returned? Doesn't the action indicate that perhaps this is an odor trail and that chemical products on the ground mark the way? We will try a similar test with the other ant species. Again, behavior shows that a crucial stimulus has been removed. How can the trail be restored? From this and other experiments, we find that the excited ant leaves a chemical trail on the ground over which it moves. Once this basic scent is laid down by the first ants that move into new ground, sensitivity to chemical stimulation permits others to follow. In the rush of two-way traffic from and to the bivwac, some booty carriers have a difficult time returning. If the ants are forced into moving on outlying trails, they often deposit their loads in booty caches. These storage areas may contain as much as a gallon of slaughtered food. Close in toward the bivwac, branching trails have disappeared, leaving only a principal trail. Over this now heavily traveled and chemically saturated path. Laden workers bring food to the brood, queen, and other inhabitants of the colony. At times, returning traffic may be dominated by unladen ants. They have simply been moving back and forth, following the scent of others on the trail. This colony is in the statory or non-migratory phase of its nomadic existence. During this phase, which lasts about 3 weeks, it remains clustered in the same sheltered site while maturing a pupil brood and undergoing early reproductive stages of a new brood. What is the nature of the population living in this bivwac? To find out, we will take a sample for later study in the laboratory. Swirling the bag rolls the ants into a ball which can be poured into an observational frame. One important section of the population is the brood. It is present in any functional colony. Here is a small part of a brood in the pupil stage attended by workers. In each colony, there is one fertile queen who is the source of the immense brood. Depending on the species, she may lay 60,000 to 200,000 eggs in each reproductive period. All workers are infertile females and all are polymorphic, that is they differ in both size and form. In this species, they range from 3 mm to 13 1/2 mm in length. Majors are the largest workers. Intermediate workers are the most numerous. They do the largest part of the effective rating. Minimms, the smallest workers, spend most of their time tending the small brood within the bivwac. Size appears to depend upon the order in which the eggs are hatched and the amount of food which larvas receive. The first larvas to begin feeding receive the most food and grow to be majors. They are characterized by their enormous heads and large sickle-shaped mandibles. These needle-sharp jaws, together with potent sting, are enough to keep most intruders away. Minimms, the last to hatch, are fed the least amount of food. The statory phase of colony behavior ends when a brood of new young workers emerges from the cocoons. Now the nomadic phase begins with daytime raids ending in nighttime immigrations to new nesting sites. Photography was possible only with batterypowered lights. In late afternoon, exodus from the bivwac begins. As the excitement mounts, ants hooked into the bivwac wall are released by the departure of other ants. As the cluster gradually breaks down, they pour from the nest and immigration begins over a raiding trail developed during the day. Late raiders returning to the bivwac are forced to change direction by the outgoing rush of traffic. These immigrants are moving their own larvas together with booty. During most of the year, successive broods contain only developing workers. The ants carry the larvas and booty in their jaws slung under their bodies. Size of the workers roughly determines the size of the loads they carry. Except that majors never seem to carry anything. They often appear to get in the way of others. At the beginning of the dry season, a bisexual brood of a great many males and a few young queens develops. Here, the workers are carrying the very large male larvas of the sexual brood. In rough sections of the trail, the column crosses on bridges made up of the bodies of clustered workers. The large ants running in the column are winged males. Development of a sexual brood of new queens and winged males leads to a splitting of the colony into two new ones, an important part of population control in army ants. This immigration will be complete only when the entire colony is settled in a new place. These ants are entering and adding to the structure of the new bivwak. Nomadic bivwaks are not sheltered but are usually out in the open. In the nomadic phase, immigration such as this go on every night for more than 2 weeks. Usually before midnight, a change in the character of the moving column indicates passage of the queen and her party. She moves slowly under her own power through the masses of workers that surge over and around her. But why do the workers respond so excitedly to her presence? Is this a chemical effect also? Let's try another laboratory test. In this frame, two circular areas have been marked. A queen has been resting on one of these areas for 1 hour. The other is our control. Now we remove the queen and her confining cover. Opening the closed tube-like tunnel permits the ants to enter. Within a short period of time, it is clear that the workers are strongly attracted to their queen's odor. In this test, note how the workers follow the queen in a circular column and how the queen responds to the chemical trail much as does a worker. Now we will return a queen to the natural environment by releasing her close to a raiding column of her colony. Slowly, a few workers are attracted to her odor. The many that follow and cluster over her indicate that the queen's odor is a central factor in colony unity. If the queen had not been returned, brood developments and consequently the nomadic function of the colony would end. Without a queen, the colony can survive only by fusion with another colony of the same species. These scenes of army ants in action illustrate how methods of field and laboratory research have been used to study problems of insect social behavior. Research on the swarm raids, for example, reveals how complex mass actions are organized through group interactions of relatively simple individuals. Perhaps the most remarkable finding about tropical army ants concerns the nomatic statory cycles shown by species adapted to surface life. A colony is nomatic when it is stimulated to a high pitch by its active brood in the laral condition. It is statory in the intervening time when the brood is in its pupil stage. As brood follows brood through the year, each colony passes through these successive cycles. There are other kinds of army ants whose patterns of life are not as well known as those you have seen. Many species which nest underground as does Labodus predator present important behavioral problems for future investigation.