Content area
Full text
INTRODUCTION
California red scale (Aonidiella aurantii), an insect pest of citrus, is under successful control by the parasitoid Aphytis melinus in many parts of the world (DeBach et al. 1971, Rosen and DeBach 1979). In inland valleys in southern California (within 40 km of the coast), the interaction appears to be well regulated in the sense that the populations persist over long periods, fluctuations in density are bounded within a narrow range, and mean population densities show no trends over long periods (e.g., DeBach 1958, DeBach et al. 1971, Reeve and Murdoch 1985).
Reeve and Murdoch (1986), Murdoch et al. (1989), and Yu et al. (1990) showed that the bark in the interior of lemon and grapefruit trees provided a partial refuge where parasitism by Aphytis was about an order of magnitude lower than in the exterior part of the tree (parasitism by Encarsia perniciosi, a species considered to play a minor role in control, was also lower in the interior). Presumably as a consequence, scale were almost 100 times more dense in the refuge. Casual observations in other groves suggested that the refuge was absent when Argentine ants were absent or rare. The ants have runs along the trunk and main branches in the interior of trees and, though they do not tend red scale, they might create the refuge in the interior by interfering with searching adult Aphytis. These observations were also consistent with DeBach's (1958) experimental evidence that red scale were more abundant in the presence of Argentine ants. An alternative or additional cause of low parasitism in the refuge could be that the bark color in the interior is not attractive to Aphytis (Gregory 1985). Scale are also smaller on average in the interior, and Aphytis is known to prefer larger scale. With scale size held constant, however, the rate of parasitism was lower in the interior (Luck and Podoler 1985, Yu 1986, Walde et al. 1989).
Reeve and Murdoch (1986) suggested that the refuge population in a lemon grove might stabilize the otherwise unstable interaction in the exterior region of the tree by leaking crawlers to the exterior at a relatively steady rate. We refer to this hereafter as the refuge hypothesis. They supported this idea by showing that temporal...