Interaction between Uroplata girardi (Coleoptera: Chrysomelidae) and Ophiomyia camarae (Diptera: Agromyzidae) on a shared host Lantana camara (Verbenaceae).

Multiple releases of insect agents intended to target a single plant pest species could result in competitive interactions that in turn might affect the community structure of the phytophagous insects. Two leaf-feeding biological control agents, Uroplata girardi Pic (Coleoptera: Chrysomelidae) and Ophiomyia camarae Spencer (Dipetera: Agromyzidae), were released against the weed Lantana camara L. (Verbenaceae) in South Africa in the 1970s and 2001, respectively. Since the population explosion of O. camarae in 2005, a decline of U. girardi populations had been observed in KwaZulu-Natal (KZN) humid coast, leading to speculation that negative interaction may be operating between the agents. The study therefore was conducted to determine the competitive effect of O. camarae on U. girardi. The study showed that 76% of O. camarae larval mines were formed on uninfested (clean) compared with only 24% formed on U. girardi-infested leaves, suggesting that the fly chose to lay more eggs on clean leaves. Almost the same number of U. girardi larval mines was formed on both O. camarae-infested and clean leaves, indicating that U. girardi females in this case oviposited indiscriminately on the two types of leaves. The survival of U. girardi was 53.8% when reared on clean leaves compared with only 14.6% survival on O. camarae-infested leaves. At the end of the sampling period, densities of U. girardi was over two times higher in single-species than in combined-species treatment. Releasing both agents together did not significantly affect O. camarae densities during the sampling period. In the field, O. camarae densities increased rapidly from spring to autumn, whereas those of U. girardi remained consistently low during the same period. The bias toward oviposition on clean leaves in O. camarae enables its larvae to avoid unfavorable encounters with U. girardi larvae, thus enhancing its development and survival. The apparent inability of U. girardi to distinguish between suitable and unsuitable leaves for oviposition could compromise the fitness of this beetle, and this could explain the suppression of U. girardi populations during summer when O. camarae populations begin to increase rapidly. This study provides evidence for an asymmetric interaction between two introduced agents, and therefore highlights the importance of conducting interaction studies on agents with extensive niche overlap before their release into the environment.