Simulation of Release of Sex-Linked Translocation Homozygotes for Population Replacemene

A deterministic computer model was built to e\'aluate \'arious release strategies of sex-linked translocation homozygotes for the purpose of replacement of a local population of Aedes aegypti (L.) in a Kenyan village. The model incorporated overlapping generations, density-dependent larval survivorship, mating competiveness of males for females, immigration of wild types, and the release of adults or pupae. The results of exercising the model indicate that the immigration of wild types can prevent the achievement of total replacement of the local population and that an equilibrium translocation frequency will be reached. The particular value of the translocation frequency depends on immigration and release rates, as well as the sex of the immigrants. In the absence of immigration, and with the exception of translocation homozygous females that have not been mated to like males, selection Deleterious effects of pesticide accumulation in the environment and development of insecticide resistance (Pal and LaChance 1974) have required the investigation of alternative methods of insect pest control. Genetic control is one of the alternatives that has been suggested (Davidson 1974). A number of techniques is included under the rubric of genetic control. One of these is to utilize chromosomal translocations (Robinson 1976) _ Curtis (1968) suggested use of translocation homozygotes that bear a gene for refractoriness to transmit a disease or a parasite. The mating of the local population with the translocated individuals would yield partially sterile translocation heterozygotes. Theoretically, if sufficient numbers of translocation homozygotes were released, they would eventually replace the local vector population with a refractory population. Such a release program was conducted in Mombasa, Kenya. The goal was to replace a local population of Aedes aegypti (L.) with a released sex-linked translocation homozygous population (Lorimer et aL 1976). As part of this project, a computer model was constructed to evaluate different release strategies that might be employed in a control program. Reported here is a description of the model and a comparison of results of several different release strategies of sex-linked translocation homozygotes simulated on the computer. MODEL.-In any mathematical model, there is the problem of satisfying 3 suggested attributes of a model, i.e., realism, precision, and generality (Levins 1966). In computer simulations, usually some degree of generality is sacrificed for more realism and 1Received for publication June 15, 1977. • Present address: Department of Biological Science., Northern lllinai. Univer.ity, DeKalb 60115. is frequency dependent. If immigration is allowed when the release is stopped, the translocation frequency declines. The rate at which it declines depends on the immigration rate. Most of the kinds of adult releases are more efficient than release of pupae. Reduced mating competitiveness of wild males for released homozygous females has little influence on the success of releases, if females have mated to homozygous males before release. It is surmised that the high intensity of larval competition decreases the influence of mating competiti\'eness in both adult and pupal releases. A decrease in adult sur\'ivorship has little effect in continous releases, if translocation females have been mated to like males before release. It does increase the time in which fixation of the translocation is achieved, if adult releases are stopped or if pupae are released. precision. This model is an extension of one developed for evaluating genetic control methods of mosquito populations by Dietz (outlined in Pal and LaChance 1974). It can be considered intermediate in its level of realism and precision. The main characteristics of the model are that it considers overlapping generations, density-dependent larval survivorship, all stages of the life cycle, migration of wild adults into an area, and the probability that mating is dependent on mating competitiveness and abundance of different classes of males. The model