Over the next 10 years, the 40-year miniaturization revolution in silicon-based microelectronics will approach its ultimate nanometer-scale limits. Molecular electronics is a potential candidate for future electronic devices. However, it is currently very difficult to predict which molecules would be suitable for fabricating such devices because the number of organic molecules that are available is enormous. In this paper, we propose that a Genetic Algorithm (GA) could be used to survey molecules to determine those which would be appropriate for practical molecular devices. The molecules that were identified in this study have a long rigid shape and a small HOMO-LUMO gap.