A South East Asian ponerine ant of the genus Leptogenys (Hym., Form.) with army ant life habits

The emigration and raiding behavior of the SE Asian ponerine ant Leptogenys sp. 1, which resembles L. mutabilis, were observed in the field (Ulu Gombak, Malaysia). The ants formed monogynous colonies that consisted of up to 52 100 workers. The bivouac sites of this species were found in leaf litter, rotten logs, ground cavities, etc., and were rarely modified by the ants. The colonies stayed in these temporary nests for several hours to 10 days; afterwards, they moved to a new nest site. The emigration distances ranged from 5–58 m. Since nest changing takes place at irregular intervals, and pupae and larvae are always present in the nest relocations of Leptogenys sp. 1, the emigration behavior is not linked to a synchronized brood development. Leptogenys sp. 1 is a nocturnal forager; in our study, up to 42 600 workers participated in each raid. The ants move forward on a broad front; behind the swarm a fan-shaped network of foraging columns converges to form a main trunk trail. A new system of foraging trails is developed in each raid. The workers search for their prey collectively; they attack and retrieve the booty together. The diet of Leptogenys sp. 1 consists mainly of arthropods. Army ant behavior is characterized by (1) formation of large monogynous colonies, (2) frequent emigrations, and (3) mass raids in which all foraging activities are carried out collectively. Since Leptogenys sp. 1 performs these typical army ant behavior patterns, this species represents the army ant ecotype. However, this species differs considerably from army ant species that have synchronized broods and huge colonies with dichthadiiform queens.

from 5-58 m. Since nest changing takes place at irregular intervals, and pupae and larvae are always present in the nest relocations of Leptogenys sp. 1, the emigration behavior is not linked to a synchronized brood development. Leptogenys sp. 1 is a nocturnal forager; in our study, up to 42 600 workers participated in each raid. The ants move forward on a broad front ; behind the swarm a fan-shaped network of foraging columns converges to form a main trunk trail. A new system of foraging trails is developed in each raid. The workers search for their prey collectively; they attack and retrieve the booty together. The diet of Introduction In the tropics of Africa and America, army ants of the subfamiUes Dorylinae and Ecitoninae are faunal elements of outstanding ecological importance. They perform impressive mass raids and colony movements. Most conspicuous are the swarmraiding epigaeic prey generalists, e.g., Dorylus (Anomma) wilverthi in Africa (Raignier and van Boven 1955) and Eciton burchelli and Labidus praedator in the neotropical region (Rettenmeyer 1963;Schneirla 1971).
Doryline ant species occur also in SE Asia, but the Dorylus species of this region are subterranean foragers, and Aenictus species prey mainly on various ant species (Wilson 1964). Swarm raiding epigaeic Dorylus species are not found in SE Asia. Gotwald (1979) hypothesized that the lack of diversity in this genus in Asia might be due to competitive exclusion by other ant species with army ant life habits that were already well established when Dorylus dispersed to this zoogeographical region. This idea is supported by investigations on Aenictus gracilis and Aenictus laeviceps in the Philippines (Schneirla and Reyes 1966). These species prey on a broader range of invertebrates and show transitions to swarm raiding behavior. Moffett (1984) (Bishop 1973).
For observations on the nocturnal ants, torch lights and head lamps were used. Two colonies were captured. The bivouac was surrounded by a square metal frame (1.5 mx 1.5 mx 0.4 m) that was dug into the ground. A bridge connected the interior of the metal frame to a metal container (1.5 m X 1.2 m xO.5 m) that was filled with a dense layer of bamboo, leaves, and branches. Intensive watering of the bivouac released an emergency nest relocation into the artificial nest. One of the captured colonies was killed with liquid chloroform and preserved in 80% ethanol.
Records of the numbers of ants participating in raids or emigrations were taken by counting the incoming and outwardbound ant traffic near the nest entrance for 1 min each in lOmin intervals. For each interval, the rate of prey-retrieving workers out of 100 returning ants was determined. Since the numbers of our observations on emigrations and raids were restricted, we used the median (x) together with ma.xima and minima instead of the arithmetical mean (x) for the investigated parameters (Sachs 1974). For the statistical analysis of the emigration routes the Raleigh-test (Batschelet 1981) was used.

Results
Nesting behavior Leptogenys sp. 1 does not exhibit any nest building activities; however, minor enlargements of the nest entrance were sometimes observed. The temporary nests (bivouacs) of Leptogenys sp. 1 were localized in subterranean cavities (n = 19), in or under decaying plant material, e.g., holes in fallen trees (« = 6), or in bamboo groves in hollow spaces in the layer of dead twigs and leaves that reach up to 1 m in depth (« = 14).
Examinations of emergency bivouacs that were formed after flooding (Fig. 1), as well as examinations of nests in large artificial arenas filled with plant material, plus the rapid killing of colonies in the field with liquid chloroform revealed the structure of the temporary nests. The ants did not form a distinct large cluster but instead were spread out over an area of about 1 to 1.5 m^. Most of the workers were distributed over the substrate in several irregularly spaced one-layer groups. Depending on the type of nest material, one of several stories with such layers of ants were formed. We also observed small clusters of ants measuring only a few centimeters in diameter. They consisted of adult and callow workers that were sitting upside-down on plant material, etc. Some of these workers held larvae in their mandibles (Fig. 2). The numerous pupae were dispersed throughout the substrate or kept in depressions in the ground where they formed several layers.

Colony size and structure
Workers. The size of our main observation colony was estimated four times during a period of Females. In 6 of the 12 emigrations of our observation colony we saw one ergatoid queen. The queen was wingless and of the same body length as the workers, but could be distinguished from the workers by her light brown cuticle, her thicker gaster, and her larger thorax (Fig. 3). During several of the observed emigrations, the queen was dis-tinctly physogastric. In 5 emigrations of other colonies observed from beginning to end we saw only a single queen in each case. We also found a queen with strongly developed ovaries in the main observation colony that was captured.
Males. The number of males that were counted during emigrations in the main observation colony varied from 52 to 746 (,v=150, « = 9). They were carried by the workers. At dusk we often observed flying males leaving their colonies. In one case we watched a male approaching the entrance of a bivouac from outside by following a foraging trail. This behavior has already been described for other Leptogenys species (Maschwitz andMuhlenberg 1973, 1975). This male may have come from an alien colony, because it was attacked by the workers; nevertheless, it went back to the trail repeatedly.
Brood. The number of pupae that were counted during the nest relocations of the main observation colony over a period of 7 weeks ranged from 6400 to 16400 (x= 1 1 200, « = 12). Since either one large larva or several small larvae or clusters of eggs were carried by a single worker, we recorded only the number of workers transporting larvae and eggs (2700-7600, X--3300, « = 12). In the captured colony we found 4580 pupae, larvae of all stages, and eggs. Only the medium-sized and large larvae (n = 3254), not the eggs and young larvae, were counted. In 21 further nest emigrations often other colonies of Leptogenys sp. 1 in different years, we also observed, in random checks, that pupae and larvae were always present at the same time, and that the number of workers transporting pupae exceeded the number of workers transporting larvae. Since pupae, larvae, and eggs were always present in about the same ratio, we conclude that in Leptogenys sp. 1 the brood is produced in an acyclic way.

Nest moving behavior
The frequent colony emigrations were preceded by at least one, sometimes two raids. The new bivouac site was always located in the area that had been raided previously. The emigrations started between 2000 and 0500 hours local time, often after the retreat of the preceding raid. At the beginning of a nest relocation, the number of workers that left the nest entrance increased. Ten to 20 min later the first pupae appeared. Before the emigration column was established, single workers carrying pupae were observed in the raiding column. These workers in; a workers carrying pupae; workers carrying larvae; below: n rate of prey-laden workers pupae were carried ventrally (Fig. 5). The ants usually moved in a single file while they were carrying brood.
The majority of the pupae were transported during the first third of the emigration; the transport of larvae began later and reached its maximum in the second third. When disturbed by heavy rain, the workers laid down the pupae in sheltered places for a while until they were able to continue with the nest relocation. After most of the brood had been transported to the new site, the number of unladen ants leaving the old nest increased.
Among the "rear-guard" many callow workers were observed. As long as ants were still leaving the old bivouac site, other workers continued to return to it. Eventually the emigration column ended abruptly. The course of a typical colony emigration is documented in Fig. 4. During the emigrations we counted 12400-31000 workers (x = 19050, 72 = 12) leaving the old bivouac site. These fluctuations seemed to be due mainly to the fact that many raiding workers did not return to the previous nesting site, but instead joined the emigration column to the new bivouac. The queen was not protected by workers when she followed the emigration trail. However, at a distance of some meters from each other, we observed several large groups of up to 100 workers that guarded the emigration trail at both sides.
They remained motionless with their heads turned towards the worker column and their mandibles opened (Fig. 5). The passing ants were checked with the antennae. below : a rate of prey-laden workers number of guests. Their number varied from colony to colony. Many species of guests followed the emigration trail during or shortly after the nest relocation. We observed at least two species of staphilinid beetles (see Note added in prooO; a lepismatid, which was sometimes transported on pupae; a reddish spider (Oonopidae); white collembolans; and females of the recently described phorid species Puliciphora rosei, which was observed to be associated with Leptogenys mutabilis (Disney 1988) and a hitherho undescribed species of Rhynchomicropteron.
Several guest species are carried on different stages of the brood apparently depending on their special residence site in the colony. Exalloniscus maschwitzi, an isopod species that is greatly adapted to its myrmecophilous life (Ferrara et al. 1987), often clings to the end of a pupa and is thereby carried to the new bivouac. We observed up to four individuals of a tiny ptiliid beetle on top of a pupa and also observed mites riding on pupae. On the larvae we found a larger dark ptiliid beetle, a whitish beetle with partly reduced wings, and several mites. A mite species was also observed on the workers. Foraging behavior Leptogenys sp. 1 forages nocturnally. The raids began at dusk and ended before or shortly after dawn. During the observation period, 1-3 raids per night (x = 2, n = l) were conducted. In every raid a new trail system was developed (Fig. 8). We registered up to 38,400 participants in each raid (16,700-38,400, x = 27,200, n = ll). Three raid ,^' Mb \3Ia Sa Fig. 8. Raid courses conducted from bivouacs 2 and 3 (no measurings were taken on nights 6 and 9, and after the nest relocations on nights 4 and 5). ( raid trail; • emigration trail) phases were distinguished: the exodus, the main raiding phase, and the retreat (Fig. 7).
Exodus. The forays started after sunset between 1920 and 1945 hours. The first activities that could be seen outside the bivouac were groups or columns of workers moving beneath the leaves. From there they expanded rapidly in several directions into the leaf litter around the bivouac. They advanced by forming pseudopodia-hke formations, which consisted of hundreds of workers, or narrow columns. The main trail on which the mass exodus occurred was estabhshed within 20-60 min (« = 5).
During the exodus, the outgoing traffic dominated  Fig. 9. Structure of a raid exemplified by a section of a developing swarm (j = swarm front ;/= fan-shaped network of trails) In the following description the data for swarm and fan are combined, because a reliable discrimination between the formations was not always possible. The swarm and fan covered up to 10.5 m in width (3.1-10.5 m, i = 6.8 m, « = 17) and 15.5 m in depth (4.6-15.5 m, x = 9.5m, «=16). Its area ranged from 16 to 58 m^(.•c = 37.5 m^, « = 14). The advancing swarm kept a main direction of progress. The maximum distance covered by the forays was 56 m (7.3-56 m, i = 24m, rt = 13). The course and advance of one swarm raid is documented in Fig. 10. Further details are given in Table 3.
We estimated that an area of approximately 300 m^was raided by the swarm every night (ex- were not considered). A detailed analysis of swarm raiding behavior will be given in a forthcoming study.

Diet
The diet of Leptogenys sp. 1 consists mainly of adult and immature arthropods, but also includes to a limited extent other invertebrates like planarians, snails, and earthworms. The largest arthropod prey individual was a scolopender (12 cm long).
Vertebrates usually escaped by running away when they got in contact with the biting and stinging ants. The only vertebrate victims were a frog (5 cm long) and a snake (15 cm long), which presumably got cornered in inadequate retreats. Non-animal food did not seem to play any role in the natural diet of Leptogenys sp. 1, although three times the ants retrieved fresh papaya seeds that were presented to them. Carbohydrate foods (bread, cookies, fruit) were not collected by the workers. Artificially offered mantids, phasmids, cicadas, diptera, and bees were taken, as well as the meat of fish or canned dog food. The ants did not attack big Achatina snails but preyed on them if the shell was destroyed. 31.1 Table 3. Leplogenys sp. 1. Size, maximum distance from bivouac, and velocity of advance of swarm raids (the data for swarm and fan area are given in combination, because a reliable distinction between these formations is not possible in each case) Night convincing explanation for the brood synchronization in L. japonica. The term "army ant", as was introduced by Wilson (1958), includes a number of ant species that live in temporary nests and attack and retrieve the prey co-operatively, e.g., Leptogenys diminuta. In this species, however, scout ants search for food solitarily and recruit a group of workers from the nest to the place where prey has been localized.
The raiding workers form a distinct group that is not connected to the bivouac by a continuous column of workers (Attygalle et al. 1988). Since the search for food is done solitarily, the foraging behavior of L. diminuta differs from the hunting strategy of real army ants in an important feature (Fletcher 1973 (Smallwood 1982). The frequent nest relocations in army ants are considered to be advantageous because the colonies are often migrating to new feeding sites (Schneirla 1971;Wilson 1971). Topoff and Mirenda (1980) demonstrated that the food supply may even have a direct influence on the emigration frequency of army ants. Franks and Fletcher (1983) analyzed the emigration behavior in Eciton burchelU. They demonstrated that similar to Leptogenys sp. 1 , the choice of a new bivouac site is not made at random. The emigrating Eciton burchelU colonies follow a certain geometrical pattern that increases the amount of unraided foraging area.
In Leptogenys sp. 1 the emigration distances are similar to the raid distances. Since we never observed that the colonies returned to their previous nest site, we can conclude that a considerable shift of their trophophoric field is achieved by the nest relocations. The migratory behavior presumably prevents the area close to the nest site from food depletion. In spite of the similarities there also are differences between the behavior of Leptogenys sp. 1 and the lifestyle of the ecitonine and doryline swarm raiders. Synchronized broods and correlated regular changes between migratory and statary phases, which are lacking in Leptogenys sp.
1, are often considered to be typical for army ants.
But we have to point out that the brood cycle of some ecitonine species is not as regular as in Eciton burchelU and E. hamatum (Rettenmeyer 1963). The ecitonine swarm raider Labidus praedator sometimes has nonsynchronous broods, and several bivouacs of this species were observed in the same site for months (Rettenmeyer 1963). Also in African driver ants {Dorylus (Anomma) spp.), emigrations are irregular (Raignier and van Boven 1955) and not correlated to the brood condition (Gotwald 1978 The queen is primarily wingless in all army ants as well as in Leptogenys sp. 1 . The morphological differences between the workers and the queen are not very pronounced in Leptogenys sp. 1, whereas in the colonies of Dorylinae and Ecitoninae, these differences between the workers and the dichthadiiform queens, which have a huge gaster, are enormous. Workers that guard the queen during emigrations (Schneirla 1971) are not found in Leptogenys sp. 1. While the workers of most ecitonine and doryline species are polymorphic, Leptogenys sp. 1 workers are monomorphic. A narrow range of size is also found in the dorylines of the genus Aenictus (Schneirla and Reyes 1966).
Recent field studies (Franks 1982;Franks and Fletcher 1983) and the data collected by Willis (1967) permit a comparison between the E. bur-chelU average swarm raid and the average foray in Leptogenys sp. 1. The raid of Leptogenys sp. 1 attains a similar width, but is much shorter and advances much slower than the foray of E. bur-chelU. The swarm raid of this ponerine ant species contains only about one-fifth of the workers that may be involved in the raids of E. burchelU. Whereas E. burchelU colonies generally produce one swarm per day, the colonies of Leptogenys sp. 1 often conduct two or more raids each night. The nightly raids of a colony of Leptogenys sp. 1 cover an area of approximately half the size of the daily foraging area of E. burchelli.

Comparison with other Leptogenys species
We began our studies on this genus with Leptogenys processionalis (L. ocelliferd) from Sri Lanka (Maschwitz andMuhlenberg 1973, 1975). This species as well has many army-ant-like habits, but also some distinct specialized characteristics. The colony size is similar to that of Leptogenys sp. 1 . The ants may change their nest frequently; however one colony was observed at the same site for at least ten weeks. In contrast to the characterization of army ant behavior already mentioned, the raiding ants leave the temporary nest to deepened trunk trails that may be used for several weeks.
In the area where our studies on Leptogenys sp. 1 were conducted, we found a total of twelve Leptogenys species, five of which have army ant characteristics. Besides Leptogenys sp. 1, these species are L. mutabilis, L. birmana, L. crassicornis, and Leptogenys sp. 2 (unidentified and possibly undescribed species resembling L. borneensis).
Though we have not yet finished our studies on the behavior of these species, differences in their ecology are obvious. L. mutabilis is an epigaeic mass raider, but contrary to Leptogenys sp. 1, the workers forage mainly beneath the leaf litter. L. crassicornis is hypogaeic, but parts of the raid columns of this small species sometimes were discovered above ground, at night.
The seven species that are not army ants in the sense already mentioned search for prey solitarily. After encountering the prey, they either recruit a group of nestmates, e.g., L. diminuta, or even attack and retrieve their prey solitarily, e.g., L. peuqueti.