Improved fallows as a soil fertility replenishing technology is widely tested by farmers in western Kenya. Sesbania sesban, Crotalaria grahamiana and Tephrosia vogelii are among the promising species planted as short duration planted fallow. However, insect pests were reported to be the major limiting factors affecting biomass production of these fallows. The objectives of this study were to 1) Determine insect pests occuring on planted fallows of Sesbania, Crotalaria and Tephrosia, 2) Evaluate the effects of prominent insect pests on biomass production of these fallow species and yields of subsequent crops, 3) Determine biology, seasonal abundance and feeding behaviour of insect persts of the promising fallow species, and, 4) Determine the scope for management of insect pests of Sesbania, Crotalaria and Tephrosia. To achieve the above objectives several studies were conducted from 1999-2001. Surveys were carried out in 7 Districts, while on-farm trials were conducted in 12 villages located in 4 Districts of western Kenya. Biomass loss assessment due to insect pests was conducted from single species fallows of S. sesban, C. grahamiana and T. vogelii and mixed stands of the two species. Growth response of C. grahamiana to defoliation was also evaluated through artificial leaf defoliation at the rate of 0, 50, 75, 100% conducted at three and six months growth stage. Feeding behaviour of Amphicallia pactolicus a defoliator of Crotalaria, damage potential of leafhopper Hilder patruelis and seasonal abundance of Sesbania defoliator Mesoplatys ochroptera are also discussed. Effects of different fallows and their management on insect pests of subsequent crops (maize and beans) was also evaluated. Results of this study showed that wide ranges of arthropods are associated with these fallows throughout the growing period and 21 species representing nine orders and 13 families were identified. Among the observed insect pests, H. patruelis was causing a major damage to C. grahamiana. Nevertheless, biomass yield averaged over several sites did not show significant difference between the insectcide treated and unprotected sub-plots. According to this study, mixing fallows showed no significant effect on insect pest abundance and biomass yield of the fallows. The study on life cycle and feeding behaviour of Crotalaria defoliator Amphicallia pactolicus showed that female moths lay eggs on the underside of Crotalaria leaves usually in batches of 80-100 and occasionally 5-10 eggs. The eggs took 7-8 days to hatch. Larval development took 24±2.4 with six larval instars and pupal period took about 20.9±4.7 days. Infestation usually starts at flowering stage ande arly instar larvae feed on leaves, third to fifty instars prefer to feed on flowers and immature pods while six instar larvae descend and feed on leaves while searching place for pupation. Two to three generations of A. pactolicus can be observed during 12 months fallow period. Out of the total food ingested more than 50% is discharged back to the soil. The larvae do not attack leguminous crops and component fallow species. Among seven Crotalaria species tested, C. incana was found resistant to A. pactolicus. Sesbania leaf beetle (Mesoplatys ochroptera) population increases during the onset of rainy seasons and declines in the absence of rainfall. Mild infestation stimulated growth response where the unprotected sub-plots yielded more biomass than the protected ones. Good compensatory growth response of C. grahamiana was noted from the artificial leaf defoliation trial conducted in two villages. Thus at Kisimba farm, greater biomass yield was obtained from C. grahamiana defoliated at the rate of 25% (16.3t/ha-1) and 50% (17.9t/ha-1) three months after sowing compared with the non-defoliated insecticide treated control (14.6t/ha-1) but differences were not significant. Although the yield from 50, 75 and 100 percent defoliation at six months after sowing was lower than the non-defoliated control, differences were not significant. Similar result was obtained from Omukato farm. Insect persts of beans such as aphid (Aphis fabae), beanflies (Ophiomyia spp.) and thrips (Megalurothrips sjostedti) were not affected significantly due to the previous season fallows.