Strumigenys behavior

Hunting / Foraging
Masuko (1984) - As Brown and Wilson (1959) stated, hunting in long-mandibulate forms relies greatly on the powerful traplike mandibular snap, whereas that of short-mandibulate forms relies on stealth in approaching prey, and tenacious grasping following the less powerful mandibular strike. In both groups the sting is used to give prey the coup de grâce. In other words, hunting tactics are more morphologically-dependent in long-mandibulate species, and more behaviorally-dependent in short-mandibulate ones. Findings concerning the specialized striking locations in the vicinity of collembolan mouthparts, and the specialized smearing behavior of several species studied, show that such behavioral adaptations in short-mandibulate strumigenites are far more elabolate than has been previously supposed. As concerns this, some phenomena of great interest were reported by Dejean (1982), all observed in Smithistruma (see Strumigenys schulzi group): 1) There might occur ambush-type hunting by the ants in the situation of high prey density. 2) Some bizzare secretion is emitted on the labral plate from its tip though the function is unknown. 3) The laboratory experiments proved that workers of Smithistruma attract collembolans, while those of Strumigenys and Serrastruma (now Strumigenys lujae group) not; he suggested a possibility of Smithistruma emitting some mimic of collembolan aggregation pheromone, but there is no further evidence yet.

body smearing behavior
Masuko (1984) - investigated feeding and what was called "body smearing behavior."

Type A behavior

Wesson and Wesson (1939), the pioneer workers on dacetine ecology, noticed an aberrant behavior in Smithistruma pergandei (=Strumigenys pergandei), and wrote that: " On 3 occasions workers were seen to bring alternately left and right forelegs to the vertex of the head, rubbing the tarsi forward and placing them on the ground. Whether this was cleaning operation or a means of transfering some substance to the substratum, or has some other significance is not known. The ant did not clean the tarsus after rubbing the head, nor did it rub any other part of its body " (p. 92). I have observed similar behavior in Pentastruma canina (=Strumigenys canina), Trichoscapa membranifera (=Strumigenys membranifera), Smithistruma incerta (=Strumigenys incerta), Labidogenys sp., and Epitritus hexamerus (=Strumigenys hexamera). It has the opposite function to that of grooming; since the ants soil their own bodies with surrounding substances. This behavior is performed almost exclusively by foragers, while foraging, particularly in highly moist situations, where workers were observed to stop and begin to scratch the substratum repeatedly with their forelegs. Following this they scrapes the dorsum of the head and the thorax, and touch the middle- and hind legs with the forelegs. The dorsum of the extended abdomen is then rubbed with backward movements of the hindlegs, or scraped with the forelegs, while in a forwardly curled position. The mandibles and the scapes are also frequently touched. In this behavior, however, the antennal funiculi and the lower mouthparts are never touched with the foreleg tarsi. This point differs greatly from the case of self-grooming. Although this smearing is directed to various body regions, most effort is clearly concentrated on the head surface. Scratching of the substratum and scraping of the body are alternately repeated restlessly, usually for several minutes, until the workers resume foraging. The ant's body surface becomes soiled by this activity, and acquires, especially, a film of water, because the smearing is mostly performed under highly moist conditions. It is of interest that this behavior is also present in dacetines with mandibles of intermediate length, such as Epitritus hexarherus and Labidogenys sp. All aspects of their smearing behavior are identical to those of Pentastruma or Trichoscapa, except that in Labidogenys the mandibles are fully opened whenever the forelegs scratch the substratum or rub the head and thorax (perhaps because the protruding mandibles would otherwise become obstacles). In contrast, Epitritus never opens the mandibles during this activity. Frequency of this behavior in Epitritus appears to be much lower than in the other species.

Type B behavior

I have observed this second type of smearing behavior in an undescribed species of the Weberistruma group (=Strumigenys) of Smithistruma (hereafter called Weberistruma sp.). This short-mandibulate species is very polyphagous and seizes collemboian, in the vicinity of the mouthparts (Mssuko, in prep.). The workers mostly do not directly scratch the substratum, but pick up minute fragments of organic material with the mandibles, and rub this material with the foreleg tarsi. After that, the ants scrape the head with the forelegs. Unlike type A behavior, this smearing appears not to be affected by the degree of moisture of the substratum. In species with type A behavior, moistening of the plaster floor of the observation nest triggers episodes of smearing, with a short latency. This phenomenon, however, has not been observed in Weberistruma sp. It has been often observed in the field that workers of this species which are about to begin foraging perform this behavior at the nest entrance, on the ground surface. I have only once collected the material which a worker held in its mandibles during this smearing in the field. It was a cylindrical mass, about 0.5 mm long and 0.2 mm diameter, and dark in color, and was probably the feces of an undetermined soil arthropod. The possible involvement of such organic material is also seen in type A smearing, in which the first scratching of the substratum is often preceded by brief antennal examination of any minute fragmentary material on the surface, followed by application of the mouthparts to it (tasting ?). Scratching is often directed to the vicinity so examined.

Neither types of smearing have been observed in the three species of Strumigenys discussed above. Although the body surface of the ants are soiled physically by their smearing activity, the degree of soiling is quite insignificant, so that the visual or tactile camouflaging effects would probably be slight at most. The fact that this behavior is performed almost exclusively by dacetines with mandibles of short or intermediate length, which must come in direct or nearly-direct contact with prey animals in order to catch them, implies that the most probable function might be to deceive prey, by concealing the ant's odor with those of surrounding organic substances.