INFLUENCE OF BEHAVIOR PATTERNS OF ADULT I NSECTS ON RESOURCE UTILIZATION AND POPULATION-DYNAMICS - A MODEL OF POPULATION-CONTROL BY 3 MAJORCOMPONENTS

Bottom-up processes (limitation by resources) and top-down processes ( limitation by antagonists) are usually considered to be the driving fo rces in the population dynamics of insects. This view implies that a p ermanent underutilization of resources by phytophagous insects or para sitoids is due to population control by natural enemies and/or disease organisms. Our contribution presents examples of phytophagous insects and parasitoids with a level of resource utilization which cannot be satisfactorily explained by the dichotomy of bottom-up and top-down pr ocesses. We start by defining the term ''available resources''. We the n bring an evaluation of the success rate of biocontrol agents introdu ced against allochthonous noxious weeds in Canada A recent status repo rt [HARRIS 1995] shows that a considerable number of phytophagous inse ct species established in Canada without their specialized natural ene mies exploit only a small proportion of the available biomass of host plants. For one of these species, the tephritid gallformer Urophora ca rdui, we show that its host selection behaviour leads to a spatial dis persion pattern of galls resulting in a strongly inverse density depen dence of resource utilization. Further examples of an underutilization of hosts are given for parasitoids attacking Urophora species in flow er heads of Centaurea and for parasitoids of aphids. Neither a bottom- up nor a direct top-down process, but rather specific behaviour patter ns are involved in these cases of resource underutilization. These pat terns may be adaptive in the sense of ''spreading the risk'', of escap ing enemies operating in a density-dependent manner or as a trade-off between host quality and host quantity. They may also be historically constrained. To account for the possible impact of behaviour, we propo se to replace the dichotomy of bottom-up and top-down processes by a t riangle scheme where changing combinations of resource limitation, lim itation by antagonists and specific behaviour patterns control populat ion densities.