Oecologia (Berl) (1981) 51:7-13 Oecologia 9 Springer-Verlag 1981 Egg Clumping, Host Plant Selection and Population Regulation in Cactoblastis cactorum (Lepidoptera)* Judith H. Myers 1, John Monro 2, and Neil Murray 3 Department of Plant Science and Institute of Animal Resource Ecology, The University of British Columbia, Vancouver, B.C. Canada 2 North Arm Road, Bowraville, NSW 2449, Australia 3 Department of Genetics and Human Variation, LaTrobe University, Bundoora, Victoria, Australia Summary. Since the successful control of prickly pear cactus by Cactoblastis cactorum in Australia, populations of plants and moths have persisted at low densities in open woodland sites. A contagious egg distribution causes overcrowding of larvae on some plants but insures low levels or no attack of other plants. This prevents extinction of plants and insects. Cactoblastis moths choose plants with characteristics which may increase the success of their larvae. Field observations and cage experi- ments indicate that large, green cactuses near previously attacked cactuses receive more eggs. Plants which are actively photosyn- thesizing are also more attractive as oviposition sites. These oviposition preferences contribute to the observed contagious egg distribution. While open woodland Opuntia and Cactoblastis populations fluctuate around an equilibrium, pasture populations may better be described by the "hide and seek" model, with the woodland populations serving as refuges. Average plant quality and varia- tion in quality are suggested as important components in the dynamics of this system. Introduction The control of prickly pear in Australia by the introduced moth, Cactoblastis eaetorum, is the greatest success achieved in the biological control of weeds (Dodd 1940). For almost 50 years since the dense cactus stands were destroyed in Queensland and New South Wales, the moth and its food plant have existed at a low equilibrium, with only occasional local resurgences of the plants being reported (White 1981). The new equilibrium was orginally described as a game of hide and seek between the plants and insects (Nicholson 1947 cited in Andrewartha and Birch 1954, and more recently in Krebs (1972), Price (1975) and Ricklefs (1976)). But long term studies by Monro (1967, unpublished data) have shown that the plants and insects in open woodlands persist at low and fluctuating densities. In these open woodland populations, the distribution of Cac- toblastis eggs on Opuntia tends to be contagious. Some plants receive too many eggs and are destroyed by the caterpillars which suffer overcrowding and starvation. Other plants are light- ly attacked and many of the caterpillars feeding on these survive to contribute to the next generation. But, even when moth densi- ties are high, a small proportion of plants escape attack for a period of time. Monro (1967) pointed out that the "oviposi- tional wastage" resulting from the contagious distribution of * This paper is dedicated to the memory of our friend Mike Sabath. moth eggs, could act as a population stabilizing mechanism. Birch (1971) discussed the role of heterogeneity in food plant quality as a population stabilizing mechanism, and Myers (1976) explored the influence of egg distribution on moth population density and stability with a computer model. Egg clumping is a population stabilizing mechanism which is important in main- taining the current density of Opuntia in Australia. While a contagious egg distribution may stabilize the insect population, it seems disadvantageous for a moth to lay eggs on plants which already have eggs. This conflict between individ- ual and population advantages of egg clumping will be resolved if moths tend to lay eggs on plants which in some way are better for their offspring. At low densities ovipositional prefer- ence will be rewarded by increased growth and/or survival of the offspring of discriminating moths. At high density, selective ovipositional behaviour may result in the overcrowding of some plants and egg wastage. We explore this situation for Cactoblastis caetorum by characterizing Opuntia stricta plants which receive eggs. We test the hypothesis that selective oviposition could be advantageous to the individual female under some conditions. Biology of Cactoblastis The control of prickly pear by Cactoblastis has been frequently discussed (Dodd 1940; Holloway 1964; Monro 1967, 1975; Mann 1969, 1970; Osmond and Monro 1981). The proper classi- fication of the shrub pears is confused and both Opuntia inermis and O. stricta occur in the literature. Since O. strieta has preci- dence, we will use that name (Murray, MS in prep.). Cactoblastis cactorum is bivoltine in most areas of Australia and adults fly in October-November and January-February. Ovi- position occurs shortly after sunset, and eggs are laid in "sticks" a continuous row perpendicular to the surface of the plant. Each egg stick contains 60-80 eggs. A female lays one or two egg sticks per night and on average produces several hundred eggs. The first instar larvae burrow into the cladode and larvae from one egg stick require approximately four cladodes for com- plete development (Monro 1967). Pupation occurs on the ground or in dried sections of the plant. Oviposition behaviour of females placed on prickly pear plants was observed. In the first stages, the female taps the plant surface with her palps and waves her antennae, while walk- ing around. When she encounters a rough surface such as an areole or a spine, she briefly explores it with the tip of her abdomen before beginning to deposit the egg stick. Flying fe- males were observed to hover near plants before landing or flying off. Therefore, female Cactoblastis seem to sample the 0029-8549/81/0051/0007/$01.40