The Entomology of Swollen Shoot of Cacao. II.—The Bionomics and Ecology of the Species involved

Abstract

A method for the routine quantitative estimation of mealybug populations on cacao trees is described and the results obtained from 24 successive monthly samples are tabulated.The quantitative results obtained are of such variability that only gross differences in population density are shown as statistically significant.From a series of analyses based on the survey data the following conclusions have been reached with regard to P. njalensis, the most important mealybug vector of swollen shoot virus :—(a) The species is almost invariably attended in the field by Crematogasterine ants, the most important ant species building protective carton tents over the mealybug colonies.(b) Crematogaster density is closely correlated with mealybug density and the ants construct their carton tents in relation to the numbers of mealybugs to be protected. Some of the arboreal Crematogasterine groups investigated are shown to be consistently associated with higher mealybug populations, and to have a more advanced tent building habit, than others.(c) Tree to tree variation in mealybug density is shown to be largely dependent on the identity of the dominant ant group.(d) No direct correlation is apparent between mealybug density and incidence of swollen shoot though populations in areas devastated by virus are usually smaller than in areas in which virus is actively spreading.(e) The factors of importance in the natural control of P. njalensis populations have been discussed in some detail. It is clear that the species is maintained by ant protection at a density level considerably above that prevailing amongst closely related pseudococcid species which are not ant attended to the same extent. At this level of “ protected steady density ” balance is maintained partly by losses during crop harvesting, and partly by swollen shoot rendering the host plants unsuitable as feeding stations. It is concluded, however, that natural enemies must, in the last analysis, be responsible for maintaining balance.(f) Since the protected density of P. njalensis is a direct result of ant associations it follows that factors tending to control the attendant ant species will have a delayed effect on mealybug density. In this respect it is believed that Oecophylla, which is strongly negatively correlated with Crematogaster spp., is of importance. Any attempts that may in the future be made to control the mealybugs by killing the ants will have to be designed as specific against the Crematogasterini and non-lethal to Oecophylla and the other large predatory ants common on cacao.