This chapter describes the application of a different genetic algorithm — Structured Genetic Algorithm (sGA) — for tracking an optimum in time-varying environments. This genetic model incorporates redundancy in chromosomal encoding of the problem space and uses a gene activation mechanism for the phenotypic expression of genomic subspaces. These features allow multiple changes to occur simultaneously, in addition to usual mixing effects of genetic operators as in standard GAs. In adapting to nonstationary environments, the extra genetic material provides a source for maintaining variability within each individual, resulting in higher steady-state genotypic diversity even with phenotypic convergence of the population in different epoch. Experimental results reported here demonstrate that sGAs can efficiently keep track of a moving optimum compared to existing genetic approaches.