EVALUATION OF THE IMPACT OF THE FRUIT-FLY PARASITOID DIACHASMIMORPHA-LONGICAUDATA (HYMENOPTERA, BRACONIDAE) ON A NONTARGET TEPHRITID, EUTRETA-XANTHOCHAETA (DIPTERA, TEPHRITIDAE)

Diachasmimorpha longicaudata (Ashmead) is an important parasitoid intr oduced to Hawaii for the control of fruit fly peats and recently has b een mass-reared in insectaries for use in augmentative biocontrol prog rams. Under field-cage and field conditions, we evaluated the potentia l negative impact of mass-reared D. longicaudata on a nontarget gall-f orming tephritid Eutreta xanthochaeta (Aldrich), which was introduced to Hawaii for control of the noxious weed Lantana camara L. In field c ages, we presented galled lantana twigs to gravid D. longicaudata in t he presence or absence of its normal rearing host, Bactrocera dorsalis (Hendel), and its food substrate (infested guava fruit) in a potted g uava tree. Regardless of the presence or absence of B. dorsalis and it s food substrate, D. longicaudata had only minimal response (visits an d probes) to galled lantana patches and E. xanthochaeta galls. As a re sult, 24-h exposure to the parasitoids (at a 5:1 ratio of parasitoids to available hosts) resulted in only 1% parasitism in the absence of B . dorsalis, and 0% parasitism in the presence of B. dorsalis. However, 26% of B. dorsalis present of B. dorsalis. However, 26% of B. dorsali s present in the cages were parasitized by released D. longicaudata. I n the field study, about 3600 gravid female D. longicaudata were relea sed in three patches of lantana plants infested with E. xanthochaeta g alls in each of two treatment sites during a 1-month period. Over a 2- month period of sampling after the first parasitoid release, 139 and 1 55 E. xanthochaeta galls successfully grew to mature (pupal) stages in parasitoid-release sites and nonrelease sited, respectively. Among th ose matured galls, 0.8% were parasitized by D. longicaudata in parasit oid release sites, and 0% were parasitized in nonrelease sites. In con trast, 37.5% of B. dorsalis were parasitized by D. longicaudata when p laced in nearby lantana patches in the parasitoid release sites, and 0 % were parasitized in nonrelease sites. Together, field and field-cage results indicate that augmentation with D. longicaudata will be unlik ely to cause significant losses of lantana gall fly populations or neg atively impact the biological control of lantana weeds. (C) 1997 Acade mic Press