The success of obligate endosymbiotic Wolbachia infections in insects is due in part to cytoplasmic incompatibility (CI), whereby Wolbachia bacteria manipulate host reproduction to promote their invasion and persistence within insect populations. The observed diversity of CI types raises the question of what the evolutionary pathways are by which a new CI type can evolve from an ancestral type. Prior evolutionary models assume that Wolbachia exists within a host individual as a clonal infection. While endosymbiotic theory predicts a general trend toward clonality, Wolbachia provides an exception in which there is selection to maintain diversity. Here, evolutionary trajectories are discussed that assume that a novel Wolbachia variant will co-exist with the original infection type within a host individual as a superinfection. Relative to prior models, this assumption relaxes requirements and allows additional pathways for the evolution of novel CI types. In addition to describing changes in the Wolbachia infection frequency associated with the hypothesized evolutionary events, the predicted impact of novel CI variants on the host population is also described. This impact, resulting from discordant evolutionary interests of symbiont and host, is discussed as a possible cause of Wolbachia loss from the host population or host population extinction. The latter is also discussed as the basis for an applied strategy for the suppression of insect pest populations. Model predictions are discussed relative to a recently published Wolbachia genome sequence and prior characterization of CI in naturally and artificially infected insects.