Eggs of Heliothis zea (Boddie) were collected twice a week from leaves of the processing tomato plants from flowering to fruit harvest. The measure of dispersion k and several other indexes indicated that distribution of the fruitworm eggs per leaf or the cumulative number of eggs per 5 leaves from the upper third of a plant was highly clumped. A sequential sampling plan for the negative binomial distribution, using a common k of 0.093 and 0.005 and 0.02 egg per leaf as the lower and upper densities for the two decision lines, respectively, was constructed. The sequential sampling plan was compared with the conventional procedure in which an action threshold of 2 eggs per 100 leaves was used. Irrespective of whether insecticides were applied on a fixed basis without scouting, or by employing the two procedures, the fruitworm damage remained the same. However, both procedures reduced insecticide applications by ca. 70%. The sequential sampling plan was better than the conventional, because the former reduced the sampling time by 20%.