Cassava green mite (CGM), Mononychellus tanajoa s.l., originating from South America, has become a serious pest in Africa, where it often occurs in mixed infestations with indigenous spider mites like Tetranychus lombardinii. The neotropical phytoseiid Neoseiulus idaeus is a candidate predator for classical biological control of CGM. Its prey preference and reproductive success on CGM and T. lombardinii were studied in the laboratory. Foraging individually on leaf discs, N. idaeus consumed CGM and T. lombardinii adult females equally in single-species populations but showed a distinct preference for CGM when both species were presented together at the same density. The predators developed and reproduced equally well on CGM and T. lombardinii.

A regional programme, involving seven national programmes and three international agencies, has been launched in East Africa to promote research on the biological control of cassava green mite, Mononychellus tanajoa (Bondar) sensu lato. The organization and objectives of the programme are described and outline results of the first 2 years of work are reported. Trials to assess losses due to mites were carried out at 10 sites, using acaricides to protect split plots of two to five varieties in each trial; at 6 sites, significant losses of tuber yield were recorded on one or more varieties. Surveys of indigenous natural enemies indicated that the predator complex is similar throughout the region and does not provide adequate green mite control. Two Neotropical phytoseiid predators have so far been introduced for local evaluation. One of these, Neoseiulus idaeus, has been released at 12 sites in three countries; recoveries were made at several sites, but no impact on the pest population was detectable. Plans for further collaborative work are discussed.

A taxonomie inventory is under way in South America to identify the major natural enemies of the cassava green mites, Mononychellus tanajoa. Special emphasis is being given to the Phytoseiidae, important predators of phytophagous mites feeding on cassava. Several insect predators and more than 40 species of Phytoseiidae have been identified in Colombia; Typhlodromalus limonicus and Neoseiulus anonymus are the most frequently found species. Selected aspects of the bionomic studies on T. limonicus and N. anonymus are presented, and methods of rearing numerous phytoseiid species are discussed. Five species are being reared at the Centro Internacional de Agricultura Tropical (CIAT) and have been sent to the International Institute of Tropical Agriculture (IITA) in Nigeria. Surveys for natural enemies will include most cassava-growing areas of Central and South America. Selection of exploration sites and timing of explorations will be guided by matching agrometeorological zones in the country of origin with those in the proposed country of introduction. Duration of dry periods, rainfall patterns and temperature are particularly important considerations in the matching.

In Africa, cassava green mite (CGM) is an exotic tetranychid, recently introduced from the Neotropics. In the absence of effective natural enemies, seasonal changes in the weather, either directly through several climatic factors or indirectly through the host plant, dominate the phenology and limit the growth rates of CGM populations. The intrinsically high rate of increase of CGM results in large populations when enhanced by high temperatures and suitable host plants during periods when evaporation exceeds precipitation. Mite numbers decline as precipitation increases and when suitable host plants are not available. The suitability of a host plant, as measured by CGM's intrinsic rate of increase, varies with drought stress, plant age, leaf age and cultivar. The introduction of effective natural enemies from CGM's area of origin is expected to dampen the seasonal abundance of this mite without changing its general phenology.

Recent developments in the assessment and collection of yield loss data due to pests are discussed. The information is needed for decision-making on resource allocation, on pest management, crop production planning and crop improvement. The difficulties in accurate loss assessment include lack of information on the distribution of pests in space and time, the difference between research and farmers yields, how much control has been used, and changes in crop resistance and in pest status. Sources of data are described and examples of losses in a number of tropical crops. The difficulties met in using losses to justify research and crop improvement projects, with particular reference to banana pests in Tanzania and cassava pests in Ghana, include the absence of reliable data, the selection of a base-line yield and the effects of other causes of loss.