“Multimodal Signals in Ant Communication”, 1999-03 ():
…Displaying ants not only walk in a stilt-like manner while raising the gaster and head, but sometimes also appear to inflate the gaster, so that the tergites are raised and the whole gaster appears considerably larger. There is also a tendency of the tournament ants to mount little stones and pebbles and display down to their opponents. In fact, the behavioral analysis of the display suggests that during encounters the contestants gauge each other’s size, and that there is a tendency among the ants to bluff, ie pretend to be larger than they really are.
From these observations we developed two models of ways in which M. mimicus may asses one another’s strength during the tournaments. Individual workers may use the rate of encounters with nest-mates and opponents (“head-counting model”) to gain a rough measure of the enemy’s strength. Alternatively, individuals may determine whether a low or high percentage of the opponents are major workers and use this information to estimate the opposing colony’s strength, since a high percentage is a reliable index of large colony size. Indeed, our investigations showed that majors are more frequently represented among tournamenting ants than among groups of foragers. Among colonies reared in the laboratory from founding queens, those younger than 4 years have a disproportionately small group of majors in the worker population.
Field experiments indicate that both assessment mechanisms are involved in inter-colony communication, and the data suggest that in particular small immature colonies rely on the “caste polling” technique, which enables them quickly to assess whether or not the opponent is a mature colony. When confronted with large workers, small colonies immediately retreat into the nest and close the nest entrance. This tactic enables small colonies to prevent larger ones from mounting a raid.
Concerning the head-counting method, our investigations revealed that it is not the entire tournamenting worker force that does the “counting”. A small group of “reconnaissance-ants” move through the tournament and gather the information. These ants are of smaller body size, and their encounter times with opponents and nest-mates are not substantially different and last only 1–3 seconds. Their trajectories in the tournament are considerably larger than those of the display ants. Individuals of this “reconnaissance-group” recruit reinforcements from the home nest, by laying chemical trails with secretions from the rectal bladder and by performing a rapid jerking display at the nest, which apparently excites nest-mates which follow the recruiting ant to the tournament site. Inspections of the condition of the fat-bodies, ovaries, and external wear and tear of the responding workers suggest that most of them are older individuals and especially ants of larger body-size remain at the tournament as display ants. Thus, the Myrmecocystus colonies communicate to neighboring colonies their resource-holding potential by summoning cohorts of large display ants to tournament sites. Colonies that are unable to match the challenge retreat and forage into other directions or wait inside the nest until the dominant neighboring colony is inactive. Indeed, I often observed in the field that foragers of large colonies stay inside the nest for days, for example when the foraging conditions are not good because it is too dry and termites, a main food source for M. mimicus, are not on the surface of the desert soil. In this “activity shadow” foragers of smaller colonies swarm out, scavenging for and hunting whatever they can find and retrieve.
The territorial tournaments may be considered one of the pinnacles in ant multimodal communication. They involve mutualistic intra-colony and manipulative inter-colony communication. By means of chemical trails and motor displays, nest-mates are summoned to the tournament site and during encounters and confrontations with other ants they use colony-specific chemical cues for recognition of nest-mates and opponents. Though it is said that the use of visual signals in ants is at least minor, and in fact not a single example has yet been solidly documented (1990), M. mimicus workers have relatively large eyes and are very good at detecting moving objects when hunting. It is therefore quite likely that vision also plays a role during tournament interaction. Certainly tactile signals appear to be very important, because the ants continuously antennate the opponent’s whole body, but especially the gaster.
View PDF: