not signi®cantly different (F = 2.20; df = 1, 30; P = 0.138). The average trap catch in the control orchard continued to increase after this to a peak of 622 on the ®fth week of spraying, while the average trap catch in the treatment orchard dropped during the week after the ®rst spray and never exceeded 100 throughout the rest of the study (Fig 1A). The average trap catch in the treatment orchard was always signi®cantly less than in the control orchard from the second week of spraying through the week after the last spray.
Fruit infestation
As of the ®rst week of spraying, the average oriental fruit ¯y infestation (including parasitoids which had parasitized eggs or larvae of oriental fruit ¯y) was not signi®cantly different in the treatment orchard (70.8 ¯ies kg À1 guava) and in the control orchard (47.7 ¯ies kg À1 guava) (F = 2.09; df = 1.16; P = 0.168). Fruit infestation subsequently dropped in both orchards, but the decrease was greater in the treatment orchard, so that the infestation there was signi®cantly less on the 3rd, 5th, 7th, 8th, 9th and 11th weeks of spraying (Fig 1B), by which time the protein bait trap catches showed that the population level in the control orchard had fallen considerably (Fig 1A). Percentage parasitization of oriental fruit ¯y by Biosteres arisanus (Sonan) was typically higher in the treatment orchard throughout the study, averaging 55.1% in the treatment orchard over the last seven spray weeks, compared to 42.1% in the control orchard (Fig 1B), suggesting that the bait sprays were not adversely affecting the population of this parasitoid.
The results presented here suggest that phloxine Bprotein bait sprays can suppress established oriental fruit ¯y populations and that phloxine B is a potential replacement for malathion in bait sprays for tephritid fruit ¯y suppression/eradication programs. However, more extensive tests are needed which limit immigration of adult ¯ies and parasitoids from surrounding unsprayed areas.