ESTIMATION OF GENETIC PARAMETERS ASSOCIATED WITH THE COMPETITIVE ABILITY OF TRIBOLIUM CASTANEUM
Abstract
When individuals coexist in a common environment, there is the possibility that they will interact. One type of interaction is competition. Earlier work in this laboratory examined the effects of natural selection for a limited food resource. It was found that the competitive abilities, of two populations of Tribolium castaneum (Pearl and Black), did respond to natural selection, but that the mechanisms responsible for the responses were different. In this investigation, the two populations of Tribolium castaneum were examined. Genetic parameters were estimated for traits important to the outcome of competition, and were evaluated in terms of direct and associate effects. The traits considered were Survival and Biomass. Each trait was examined in two environments; Mixed and Separate cultures. In Mixed cultures, the two populations (Pearl and Black), were evaluated in competition with a third Tester population ("+" foundation). The Pearl and Black populations were structured in such a manner to allow the estimation of genetic parameters (heritability and genetic correlations). The genetic parameters were then used to predict direct and correlated responses to selection. Selection responses were predicted based on a multiple trait model, which accounted for both the effect of the environment and associate effects. The selection responses (predicted), based on the multiple trait model, were compared to those based on the classical model (not accounting for associate effects). It was found that predicted selection responses, based on the multiple trait model, differed from the classical model. However, the responses based on the multiple trait model were subject to large sampling variances. Despite the large sampling variance, the multiple trait model represents a more realistic representation of the biological situation, and was used to account for past responses to selection.
Degree
Ph.D.
Subject Area
Genetics
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.