Tests were made to determine the efficacy of a new bait (Maxforce; Combat) against established populations of B. germanica. The bait, provided for tests by American Cyanamid Company, consisted of semisolid food substance containing 1.65%Twt/wt) amidTnohydrozone (CL217.300), the active ingredient in Amdro fire ant bait. Bait was presented in childproof plastic stations (5.1 × 5.1 cm × 10 mm tall) in infested aaartments near Los Angeles. Each apartment was about 37 m2 (400 ft2). Most cockroaches were in the kitchens, which were about 7.9m2. The performance of bait was monitored with 3 Black Flag sticky traps or 0.95-litre jar traps baited with bread. Traps were left under the kitchen sink and behind the stove and refrigerator for 7 days. Control was calculated from the difference between pre- and posttreat trap counts. Bait was placed in cabinets, behind appliances, and close to sites suspected of harboring cockroaches. Most bait was used in kitchens but additional stations were placed in other rooms in a few instances where large numbers of cockroaches were observed. During the test, stations were replaced only at sites in apartments where all the bait had been consumed. A simultaneous comparative test with a high rate of a commercial 0.5% Dursban bait was made in apartments at the same study site.