You are not currently logged in.
Access JSTOR through your library or other institution:
Ant Lion Zones: Causes of High-Density Predator Aggregations
Nicholas J. Gotelli
Vol. 74, No. 1 (Jan., 1993), pp. 226-237
Stable URL: http://www.jstor.org/stable/1939517
Page Count: 12
Preview not available
In this study, I describe the causes of persistent small-scale zonation in a terrestrial arthropod community. In central Oklahoma, two species of predaceous, pit-building ant lion larvae (Myrmeleon immaculatus and M. crudelis) coexist in a narrow, high-density (>50 animals/m^2) zone at the base of sandstone cliffs, where they are sheltered from rain and afternoon sun. Although larval movement is frequent and abundance within the zone varies seasonally, ant lions are rarely found >2 m from the cliff base. I used field experiments and observations to test four hypotheses that could explain the maintenance of the ant lion zone: (1) zonation because of habitat availability; (2) zonation because of food availability; (3) zonation because of increased surface temperatures in exposed microhabitats; (4) zonation because of increased soil moisture and disturbance in exposed microhabitats. Ant lions are probably not limited to the cliff base by the presence of rocks or vegetation, because there was no correlation between ant lion density and available habitat. The availability of prey also does not appear to limit ant lion distribution: ant abundance in pitfall traps was significantly higher outside the ant lion zone than within. Outside the ant lion zone, soil surface temperatures reach lethal levels in the afternoon sun. However, third-instar larvae of M. immaculatus transplanted in dry, sifted soil suffered little mortality over a 4-d period. Larvae transplanted into exposed sites gained significantly more mass than larvae transplanted into the ant lion zone, probably because of differences in food availability. Small-scale manipulators within the ant lion zone revealed that ant lion numbers decreased significantly over a 5-d period in response to both moisture (misting) and disturbance (pit-filling). In a recolonization experiment, the recovery time of patches exposed to a single @'rainfall@' treatment was extremely long: defaunated @'rainfall@' patches did not achieve the pit densities of defaunated control patches until after 84 d. Rainfall limits ant lion distribution through an interaction with temperature. Transplant experiments on natural substrata revealed the nature of this interaction. Within the ant lion zone, the soil remains dry and ant lion larvae bury themselves more rapidly in the sun in the shade. Outside the ant lion zone, rainfall leads to the formation of a persistent soil crust. Larvae are unable to penetrate this crust, and they succumb to high temperatures in the sun. Because both species of ant lions are restricted to a narrow spatial zone, the probability of interspecific and intraspecific interactions (including competition, predation, and cannibalism) is increased. The abundances of M. crudelis and M. immaculatus were negatively correlated in quadrats from the center of the ant lion zone. Across the ant lion zone, the relative abundance of the two species also differed significantly, although the differences were not consistent among sites. Because of overriding abiotic constraints, neither species can achieve an ideal free distribution with respect to food resources. High-density predator aggregations may also affect prey community structure. Arthropod abundance was low within the ant lion zone, perhaps because of direct predation by ant lions or predator avoidance behavior by prey. Although the ant lion zone is caused by the interaction of abiotic factors, high-density predator aggregations have important biotic consequences for the population dynamics of the predators and the spatial distribution of their prey.
Ecology © 1993 Wiley