In two small plot tests, populations of plant bugs and bollworms were too low to determine differences among treatments. However, in test 1 (12 rows X 30 ft.), after two applications, Steward at 0.09 or 0.11 lbs. a.i. caused ca. 6- and 25- fold reductions, respectively, in ant numbers and Karate Z at 0.03 lbs. caused a 12-fold reduction compared with the untreated plots. There were no significant differences between untreated plots and those treated with Tracer at 0.08 lbs. a.i. In a second test (plots 16 rows X 40 ft.) in which two applications of Steward at 0.09 and 0.18 lbs. a.i. were compared with an untreated check, ant populations were reduced significantly and substantially by both rates after one and two applications. Although differences in numbers of geocorids were not significant after the first application, the high rate of Steward had significantly lower numbers of adults than did untreated plots following a second application. Additional studies are needed to assess the impact of Steward on beneficials, particularly in areas where ants and geocorids are important predators.