Discrete frequency distributions, obtained from count data onSolenopsis invictaBuren,Aphisspp.,Pseudatomoscelis seriatus(Reuter) adults and nymphs.Orius insidiosus(Say).Nabisspp., andGeocorisspp., in 1973 and 1974, were fitted to seven mathematical frequency distribution models. The negative binomial distribution best described the dispersion pattern of these insects in both years, indicating an aggregated, nonrandom dispersion pattern. The parameter k of the negative binomial distribution was used to quantify the aggregation of these insects. The maximum likelihood estimate and the moment estimate were two methods used to estimate the parameter k. A common k was calculated for each insect studied, to determine whether the estimated k values were common to all sampling dates over a season. As indicated by common k values,Aphisspp. andS. invictawere highly aggregated.P. seriatusnymphs were more aggregated than the adults. Among the predators,Nabisspp. were more aggregated thanO. insidiosus;andGeocorisspp. were the least aggregated. Taylor's power law was fitted to the variance and mean density values for each insect. The power function well described the relation between the variance and mean density value for each insect