The most important species of Pseudoeoccid transmitting the swollen-shoot virus disease in the Gold Coast is Pseudococcus njalensis Laing. It is almost always attended by ants of the genus Crematogaster. These ants usually build their nests in the cavities and galleries that have already been excavated by wood-boring insects in the dead branches of cacao trees. The only apparent advantage of this association to the ants is to imbibe the honeydew secreted from the anal orifice of the mealybugs; if this is allowed to accumulate it becomes a medium for bacteria and fungi which seems to kill the mealybugs eventually.

Attempts were made to break the link between the mealybugs and ants by spraying six cacao trees, not in contact with each other or any other trees, each with 3·5 litres of 0·2 per cent. DDT emulsion, twice at two-week intervals. All dead branches containing nests of ants were cut out, and a band of grease painted, on the trunk, one foot above the soil level to prevent the ants climbing up. Four weeks after the first application of spray, the population of mealybugs was brought down to 1·2 per cent, of its size before treatment. Unfortunately locating the ant nests, especially in the crevices and under the bark, is very difficult. Experiments were therefore carried out in which the trees were sprayed without previously eliminating all the sources of ants. The results were unsatisfactory, the size of the mealybug population, compared with that before treatment, after four, six and eight weeks was 16·9, 39·4 and 92·3 per cent., respectively, in the case of 0·2 per cent. DDT emulsion, and 11·6, 64·3 and 58·8 per cent., respectively, in the case of a treatment consisting of 3·5 litres per tree of a mixture of 0·2 per cent. DDT and 0·02 per cent, parathion active material. Cutting off the dead branches containing ant nests, followed by either painting the cut end of the dead branches with DDT (16·7 per cent, emulsion) or painting the trunk with a band of the same emulsion also gave poor results.

The unsatisfactory results obtained by spraying are attributed to the fact i that only the ants that happen to be exposed are affected by the application of ithe insecticide; the immature stages inside the nests are not touched.

The efficiency of dimefox when applied to the soil at 0·8 gm. active material per inch of tree girth was not increased when the ants had previously been partially eliminated.

Treated cacao trees could not in practice be completely isolated from forest jtrees and climbers containing nests of Crematogaster ants and these provide a continuous source of large populations. It is, therefore, considered that the icomplete elimination of ants is not possible.

It was also found that there is a relationship between the population of mealybugs and ants. Trees were usually free from mealybugs if no ants were found on them. The examination of isolated cacao trees with different sizes of dead branches containing nests of ants suggests that the number of mealybugs 38 in proportion to the size of the nests.