SummaryThe functional response of Cyrtorhinus lividipennis feeding on brown planthopper (BPH) and green leafhopper (GLH) eggs was found to be Holling's Type II. The Random Predator Equation fitted the data satisfactorily. Using Manly's preference index, α, both the male and female C. lividipennis were found to prefer BPH eggs. There was, however, no evidence of switching and the ratio of the respective searching efficiencies could account for the preference.

Functional responses of the wolf spider, Pardosa pseudoannulata (Boesenberg et Strand) attacking the rice brown planthopper, Nilaparvata lugens (Stgd.), and the mirid predator Cyrtorhinus lividipennis Reuter were both those of Holling Type II. The attack rate was higher and handling time lower for C. lividipennis. However, when caged with the two prey, the wolf spider showed a significant preference for N. lugens at a lower prey proportion. Proportions of prey attacked were significantly different from the expected ratios of prey available as well as from the predicted preferences derived from the functional response parameters. As proportions of N. lugens attacked changed from greater to less than expected as the proportions of N. lugens available increased, a "reverse switch" behaviour seems to be evident.

Comparative field observations of pest incidence on five selected varieties each of Japonica, Indica, and Japonica × Indica rice types indicated that tungro incidence was relatively higher on Japonica varieties, while stem borer incidence at heading stage was higher on Indica varieties. No differences in the relative abundance of other pests on these rice types were observed. Insecticide treatment and caging of plants reduced pest incidence on all varieties. Plant growth and yield of insecticide-treated plants were significantly higher than unprotected plants. Caging reduced the number of tillers but the yields remained higher than unprotected plants. Yield loss estimated by the insecticide protection method was higher at 39% than that by the caging method at 25% overall. There was a significant positive correlation between yield loss estimated by the two methods in Indica and Japonica × Indica varieties and no such relationship in Japonica varieties.