Enslaved ants: not as helpless as they were thought to be.

Slavery in ants involves robbing of brood of host ant species and rearing captured individuals in the enslaver's nest. Whereas slaves of facultative slave-makers increase the workforce of the colony, in obligate slave-makers presence of slaves is vital for colony survival. Until recently, it was generally believed that enslaved workers act solely for the benefit of their social parasite and are wholly lost for their own colony and population. However, evidence that slaves may act also in favour of their own maternal population by engaging in various forms of the so-called slave rebellions is already quite extensive and may be found in both old and recent myrmecological literature, although, unfortunately, these data are often neglected or overlooked. They may be classified into four categories: (1) acts of physical aggression directed by slaves to slave-makers, (2) attempts of slaves to reproduce within a slave-maker colony, (3) 'sabotage', i.e. activities of slaves leading to weakening of the slave-maker colony and population, and (4) slave emancipation, i.e. partial or complete self-liberation of slaves from slave-maker colonies. In this review, we present and discuss all these diverse (often interrelated) expressions of slave opposition to their enslavers, focussing our discussion on both proximate and evolutionary causation of the discussed phenomena. We also indicate some open questions which remain to be answered by future research.

Octopamine reverses the isolation-induced increase in trophallaxis in the carpenter ant Camponotus fellah.

Social deprivation is an unusual situation for ants that normally maintain continuous contact with their nestmates. When a worker was experimentally isolated for 5 days and then reunited with a nestmate, she engaged in prolonged trophallaxis. It is suggested that trophallaxis allows her to restore a social bond with her nestmates and to re-integrate into the colony, particularly via the exchange of colony-specific hydrocarbons. Octopamine reduced trophallaxis in these workers as well as hydrocarbon transfer between nestmates, but not hydrocarbon biosynthesis. Administration of serotonin to such 5-day-isolated ants had no effect on the percentage of trophallaxis. Administration of phentolamine alone, an octopamine antagonist, had no effect, but when co-administrated with octopamine it reduced the effect of octopamine alone and restored trophallaxis to control levels. Moreover, the observed effect of octopamine was not due to a non-specific effect on locomotor activity. Therefore, we hypothesise that octopamine mediates behaviour patterns linked to social bonding, such as trophallaxis. On the basis of an analogy with the role of norepinephrine in vertebrates, we suggest that the levels of octopamine in the brain of socially deprived ants may decrease, together with a concomitant increase in their urge to perform trophallaxis and to experience social contacts. Octopamine administration may reduce this social deprivation effect, and octopamine could therefore be regarded as being partly responsible for the social cohesion between nestmates in ant colonies.

Digging behaviour and responses to photic and gravitational cues as elements of escape behaviour of bumblebees.

We have investigated escape behaviour of workers of two bumblebee species, Bombus terrestris and B. pascuorum, when confined to test tubes plugged with soil and either exposed to sunlight or kept in darkness. In both these situations B. terrestris performed better (i.e. escaped after a shorter time) than B. pascuorum. B. terrestris (but not B. pascuorum) also performed better in darkness than in tubes exposed to sunlight. This implies that in both situations B. terrestris showed higher readiness to dig than B. pascuorum, and that in tubes exposed to sunlight only B. terrestris showed high readiness to display photopositive behaviour as well. B. pascuorum displayed, however, photopositive behaviour in another escape situation: when released in a dark room in front of a vertical array of four sources of white light. In that situation, B. pascuorum also displayed the tendency to fly upwards, based most probably on responses to gravitational cues.

The effect of past and present group size on responses to prey in the ant Formica polyctena Först.

We investigated the responses to insect prey (dead houseflies) in 24 "derivative groups" of workers of the ant Formica polyctena created by taking sets of 25 workers out of nine larger "initial groups" kept in laboratory without queens and brood during the preceding five months. In the derivative groups the ants ceased to retrieve flies to their nests after a period ranging from few days to several weeks. The duration of that period did not depend on the present size of the derivative group (decreasing as a result of worker mortality), but was positively correlated with the estimated size of the initial group of the tested ants. The readiness to display venom spraying was higher in smaller derivative groups. These data demonstrate that responses of F. polyctena to insect prey are strongly influenced both by the present and the past size of their group.

Laboratory observations of nuptial flights of the ant Polyrhachis laboriosa.

Ethology of Polyrhachis laboriosa, an ant species from equatorial Africa, is little known. No field observation of a nuptial flight of these ants was ever made. We describe two nuptial flights observed in a laboratory colony of P. laboriosa at a 3 days interval. They both occurred in the morning while the nest was kept in near darkness (less than 2 lux of daylight). Flying activity of the alates was suppressed within 1 h by their exposure to daylight of about 140 lux, and within several minutes by their exposure to a lamp emitting white light of 3,000 lux and acting as a source of heat. On the day following the first flight the alates and the workers showed exceptionally high level of mutual grooming. The alates, in particular the males, were transported by workers to the brood chambers whenever they strayed outside and after the nuptial flights.

Factors influencing the responses to nest damage in the African weaver ant. Oecophylla longinoda (Latreille).

Responses of the African weaver ants (Oecophylla longinoda) to nest damages were studied in the field in Nigeria. During the wet season, the ants responded to nest damages almost unexceptionnally by a quick onset of nest-repairing behaviour. The latencies to the start of nest-repairing activities (LN) did not depend on the size of the damage, but they were significantly shorter during the night, and positively correlated with ambient temperature. During the dry season, the ants responded to large nest damages mainly by abandonning the nest. In the case of medium size damages, the onset of nest-repairing behaviour was equally rapid as during the wet season, but in the case of small damages it was sometimes greatly delayed (up to three hours). The values of LN did not differ between the nighttime and the remaining times of the day, and they were not correlated with ambient temperature.

Biting behaviour as a tactic of escape in two bumblebee species with different nesting habits, Bombus terrestris L. and B. pascuorum Scopoli (Hymenoptera: Apidae).

Workers of two bumblebee species, Bombus terrestris L. and B. pascuorum Scopoli, had to find the way of escape out of a test tube closed with a paper membrane stretched against its open end. Nearly all of the tested individuals solved successfully that task by tearing a hole in the membrane closing the tube. However, their escape behaviour showed significant interspecific differences. B. terrestris started biting the membrane sooner than B. pascuorum, and they were biting it more persistently. These behavioural differences matched well the differences in the nesting ecology of these two species. Whereas B. pascuorum is a surface-nesting species, B. terrestris nest in underground cavities connected with the outside world by long tunnels. B. terrestris are thus more likely to be well adapted to deal with obstacles obstructing their way, and/or to have more experience in removing them. Neither the efficiency of biting behaviour as a tactic of escape, nor the total test time differed significantly between the tested species.

Some simple techniques helpful in ant research.

Eight simple techniques which may be helpful in ant research are presented. They facilitate: (1) collecting ground-nesting ants; (2) prompting ants to go out of their nest chambers; (3) introducing ants into a test tube; (4) recapturing ants during their mass escape in the laboratory; (5) keeping tidy foraging areas of artificial ant nests; (6) keeping high level of air humidity in foraging areas of artificial ant nests; (7) providing ants with water of improved quality; (8) cleaning artificial ant nests carved in plaster of Paris.

Digging as tactics of escape in two bumblebee species with different nesting ecology: Bombus terrestris L. and B. pascuorum Scopoli.

Hymenoptera respond to confinement by vigorous and persistent escape behavior. In a field study, workers of two bumblebee species, Bombus terrestris L. and B. pascuorum Scop., were tested in glass tubes plugged by soil at the open end, and with their other, closed end oriented towards the-sunlight, so that the bees could alternate between two escape tactics: photopositive behavior and digging behavior. The bees of both species proved to be able to sdve such an escape task, i.e. to dig their way out of the tube almost without exceptions. However, as expected, workers of B. terrestris, a species nesting in underground cavities connected with the outside world by long tunnels, performed better than workers of B. pascuorum, a surface-nesting species. The workers of B. terrestris started to dig earlier, were digging more persistently and more efficiently, and, consequently, escaped out the test tubes earlier than the workers of B. pascuorum. High intraspecific variability in all parameters characterizing digging behavior of the bumblebees was also recorded.