To cite this page, please use the following:
· For print: . Accessed
· For web:
Two studies have been carried out on the behavior of P. amabilis, based on several captive colonies from La Selva Biologica Station in Costa Rica (Hoelldobler and Wilson 1986, Hoelldobler et al. 1992). The following account is based on those studies:
Colonies are polydomous, occupying many small chambers in dead wood fragments. They may contain over 700 workers, and 1 or 2 reproductive queens (active ovaries, filled spermatheca). Workers all appear capable of producing trophic eggs, which may be the main food of the queen. Workers attack live prey and transport it back to the larvae. When presented with a "cafeteria" of diverse prey types, they strongly prefer campodeid diplurans, but will occasionally take other small arthropods. Workers produce a trail pheromone from specialized glands in the hind basitarsus. They exhibit a characteristic behavior, a rapid up and down vibration, which appears to excite or attract other workers, and they drag their hind legs along the surface, apparently leaving a pheromone trail. These behaviors are used to recruit nestmates when prey is discovered. Workers exhibit a simple temporal caste polyethism, in which young workers attend primarily to brood care, while older workers undertake a broader range of behaviors, including brood care, tending the queen, and foraging. Workers have a "wall-papering" behavior, in which they use cocoon fragments to cover the walls of chambers containing pupae, a behavior thought to aid in keeping the pupal chamber dry. Trophallaxis and adult transport have not been observed, and there appears to be no alarm pheromone (Hoelldobler and Wilson 1986, Hoelldobler et al. 1992). [unclear if these studies were of P. amabilis or P. antillana.]
This is a very common species in wet forested habitats. I most often encounter it in Winkler samples of sifted litter from the forest floor, but it is also quite easy to encounter nests in rotten wood and under loose bark. Colonies can be large, with many hundreds of workers.
I once observed alate queens flying in a downpour in Corcovado National Park; they were falling down my shirt and stinging my neck. Colony founding may be pleometrotic: I lifted the loose bark from a rotten log in Penas Blancas, and found three different pairs of dealate queens, each pair separated from the next by about 30cm. The queen pairs were in small chambers and had small brood piles. In the 1986 Hoelldobler and Wilson study they found single queens in the two colonies they examined in detail, but in the 1992 Hoelldobler et al. study they found two reproductive queens in a colony. Could it be that this species regularly has two queens? The single queens observed by Hoelldobler and Wilson could be due to undersampling; locating all portions of a polydomous colony in the leaf litter can never be achieved with certainty.
Borgmeier, T. 1949. Formigas novas ou pouco conhecidas de Costa Rica e da Argentina (Hymenoptera, Formicidae). Rev. Bras. Biol. 9:201-210.
Brown, W. L., Jr. 1960. Contributions toward a reclassification of the Formicidae. III. Tribe Amblyoponini (Hymenoptera). Bull. Mus. Comp. Zool. 122:143-230.
Forel, A. 1909. Ameisen aus Guatemala usw., Paraguay und Argentinien (Hym.). Dtsch. Entomol. Z. 1909:239-269.
Hoelldobler, B., Obermayer, M., Wilson, E. O. 1992. Communication in the primitive cryptobiotic ant Prionopelta amabilis (Hymenoptera: Formicidae). J. Comp. Physiol. A. Sens. Neural Behav. Physiol. 171:9-16.
Hoelldobler, B., Wilson, E. O. 1986. Ecology and behavior of the primitive cryptobiotic ant Prionopelta amabilis (Hymenoptera: Formicidae). Insectes Soc. 33:45-58.
Found most commonly in these habitats: 29 times found in lowland rainforest, 14 times found in tropical rainforest, 15 times found in ridgetop cloud forest, isolated peak with oak trees, 10 times found in wet forest, 6 times found in mature wet forest, 10 times found in montane wet forest, along ridge descending to major river, a mix of old 2ndary forest on one slope and more mature forest on the other, 8 times found in montane wet forest, along ridge leading to stream, oak trees present, probably mature or old 2nd growth with relict trees, near pasture, 6 times found in mature rainforest, edge of forest near pasture and agricultural land, steep rocky terrain, 5 times found in rainforest, 2 times found in tropical wet forest, ...
Found most commonly in these microhabitats: 104 times ex sifted leaf litter, 1 times ex rotten log, 2 times under bark of fallen tree, 3 times pan trap, 1 times nest in dead wood, 1 times Wet forest. Nest in dead wood., 2 times wet forest, in dead wood, 1 times wet forest litter, 2 times sifted litter (leaf mold, rotten wood), 2 times in rotten wood, 1 times in dead wood, ...
Collected most commonly using these methods: 73 times miniWinkler, 23 times maxiWinkler, 16 times search, 8 times Winkler, 2 times hand collecting, 3 times Pan Trap, 2 times Berlese, 1 times FIT, 1 times litter sample, 1 times baiting, 1 times Ex rotten log, ...
Elevations: collected from 5 - 1500 meters, 557 meters average