Rastrococcus invadens
(fruit tree mealybug)

R. invadens does not seem to cause primary feeding damage to its tree host plants. However, the accumulation of honeydew and, shortly thereafter, development of sooty mould affects the photosynthetic capacity of the plant. Heavily affected plant parts stop growing, and often no new leaves or flowers are produced. In severely affected areas of West Africa, mango production was stopped altogether (Williams, 1986; Agounké et al., 1988; Bokonon-Ganta and Neuenschwander, 1995) before biological control substantially lowered mealybug populations.

In the original description by Williams (1986) about 12 plant hosts are mentioned. For Africa, Agounké et al. (1988) listed 45 species of host plants from 22 families attacked by, or harbouring populations of, R. invadens in West Africa, and Biassangama et al. (1991) listed 23 species from Central Africa. Since then, a total of over 100 host-plant species have been found in Africa, particularly where populations of this insect are abundant on the primary host, mango. Not all these hosts could sustain populations of R. invadens.

Biology

R. invadens females produce first-instar larvae, which, under field conditions in tropical Africa, moult within 10-12 days into second instars. The second instar lasts 7-8.5 days, and slight differences can be observed between the sexes. Third-instar males form a cocoon and go through to a fourth instar over 8-11 days; third-instar females take 6.5-8.5 days before moulting to adults. Overall, males take ca 28-31 days from hatching to last moult. The short-lived adult males are capable of mating upon emergence. Females take 25-27 days from hatching to adult emergence. The prereproductive period of the females lasts for 17-18 days. Females survived up to 225 days and laid eggs up to about day 200. These and further life-table parameters are given by Boavida and Neuenschwander (1995a). For R. invadens kept in the laboratory or the glasshouse, longer development times and a shorter reproductive period were found (Willink and Moore, 1988). Up to almost 200 first instars were produced on average during the life time of one female.

Population dynamics

In West Africa, R. invadens has been shown to disperse very well, most likely by humans transporting seedlings from nurseries (Boussienguet and Herren, 1991; see map in Neuenschwander et al., 1994). Population densities were usually higher on young than on old leaves (Boavida and Neuenschwander, 1995a) and differed markedly between individual mango trees. On highly infested mango trees, the pre-reproductive period was shorter and the total offspring production higher than on less-infested trees (Boavida and Neuenschwander, 1995b), indicating the importance of plant genotype on mealybug size and survival. Similarly, large differences in mealybug population levels between different mango trees were reported by Narasimham and Chacko (1991).

Population peaks occurred irregularly, but mainly in the wet season (Agricola et al., 1989; Boavida and Neuenschwander, 1995a), though they often seemed to be more influenced by the plant host than by weather (Matokot et al., 1992). The proportion of male mealybugs showed large and unexplained fluctuations, independently of population density (Boavida and Neuenschwander, 1995a).

In three studies in Africa, namely in Togo, Congo and Benin, population dynamics were heavily influenced by classical biological control. Pest populations crashed within 1-2 years, and local extinction was sometimes observed.

Predators of R. invadens are listed by Agounké et al. (1988), Matokot et al. (1992) and Boavida and Neuenschwander (1995a). Despite this richness in the indigenous fauna, control was eventually achieved only by introduction of Gyranusoidea tebygi, assisted in some foci, mainly towns in Benin, by Anagyrus mangicola. Both parasitoids were described by Noyes (1988, 1990) only after R. invadens had been recognised as a separate species. See Control for further information.

Though G. tebygi is attacked by several indigenous hyperparasitoids in Africa (Agricola and Fischer, 1991; Biassangama et al., 1991; Matokot et al., 1992; Moore and Cross, 1992; Boavida et al., 1995a, b), biological control of R. invadens does not seem to be severely affected. Similarly, levels of parasitism by G. tebygi were reduced in the laboratory by the pathogen Hirsutella cryptosclerotium (Fernández-García et al., 1990), but overall mortality of the mealybug was greater when both agents were acting together (Akalach et al., 1992).

Due to the variable regulations around (de)registration of pesticides, your national list of registered pesticides or relevant authority should be consulted to determine which products are legally allowed for use in your country when considering chemical control. Pesticides should always be used in a lawful manner, consistent with the product's label.

Biological control by Gyranusoidea tebygi and Anagyrus mangicola has led to satisfactory control of the mango mealybug in Africa. No other interventions were needed, although it must be stressed that, particularly in urban environments, more than 1 year may be needed before low host equilibria are established.

The establishment and spread of G. tebygi in Togo (six releases, Agricola et al., 1989) in Congo (where it appeared without deliberate releases, Matokot et al., 1992), and of the two parasitoids in much of West Africa (a total of 42 releases of G. tebygi and 22 releases of A. mangicola, Neuenschwander, 1989; Neuenschwander et al., 1994) are well documented. The impact, mostly of G. tebygi, was demonstrated by population dynamics studies (Agricola et al., 1989 in Togo; Matokot et al., 1992 in Congo; Boavida and Neuenschwander, 1995a in Benin, Pitan et al., 2000 in Nigeria) and by a parasitoid exclusion experiment by means of sleeves (Boavida et al., 1995b). Similarly, quantitative survey results covering all infested zones in Benin indicated a reduction of R. invadens following the introduction of G. tebygi, a corresponding reduction in sooty mould cover, and an increase in the number of new leaves (Bokonon-Ganta and Neuenschwander, 1995). R. invadens populations invariably crashed during the year following the first introduction or observation of G. tebygi. As the mealybug population was reduced, the number of host plants attacked was also diminished (Boussienguet and Mouloungou, 1993). A first attempt at economic impact analysis was made by Vögele et al. (1991). Subsequently, Bokonon-Ganta et al. (2002) estimated the benefit-cost ratio of the biocontrol project as 145:1 for Benin alone.

In addition to these field studies, both parasitoids were also studied in detail in the laboratory in order to understand their interactions with the host and with each other (Willink and Moore, 1988; Moore and Cross, 1992, 1993; Cross and Moore, 1992; Bokonon-Ganta et al., 1995, 1996; Boavida et al., 1995a). Though there is some niche overlap, the two parasitoids can coexist on the same host population. Initial fears, derived from an early simulation model (Godfray and Waage, 1991), that the addition of A. mangicola to the system could be harmful, were thereby allayed by these studies.