Abstract Spray trials were conducted in Kent and New Castle Counties, DE, and Salem and Cumberland Counties, NJ. Treatments consisted of oil-based isolates of Bacillus thuringiensis formulated as Condor OF and Dipel 8 L diluted with water, Condor OF applied undiluted and an untreated control. Treatments were replicated five times in a randomized complete block. Woodlots were 10-84 acreas in size and consisted of isolated stands. Fifteen 0.025-acre fixed radius plots were established within each woodlot where treatment effects were measured. A Grumman Ag-Cat 600 aircraft equipped with eight AU-5000 Micronair rotary atomizers on a standard spray boom was used to apply the treatments. All B. thuringiensis formulations were applied at 20 billion international units (BIU)/acre (40 oz formulation/acre). For the diluted applications, Condor and Dipel were applied at a volumetric rate of 128 oz/acre including 2% (v/v) Bond (Loveland Industries, Inc.) as a sticker. The aircraft was operated at an airspeed of 90 mph on a 75 ft swath width, about 50 ft above the forest canopy. The pump pressure was 40 psi and the variable restrictor unit on the nozzles was set at 9 and 5 for the diluted and undiluted applications, respectively. The blade angle on the nozzles nearest the fuselage was set at 65 and 35°, and 70 and 35° for the other six nozzles on the boom, for the diluted and undiluted applications, respectively. Woodlots were treated on 22 and 27 May in Delaware and New Jersey, respectively, when GM was predominately 3rd instar. Pre- (Spring) and postspray (Fall) egg mass counts were made within each of the 15 subplots. GM larvae were counted under burlap bands placed on one dominant preferred tree 25 ft outside each 0.025- acre plot. Estimates of feeding damage were based on examination of each burlapped tree and were rated in one of seven classes: 1, 0%; 2, 1-5%; 3, 6-10%; 4, 11-25%; 5, 26-50%; 6, 51-75%; and 7, >75.