The Mediterranean fruit fly, Ceratitis capitata (Wiedemann) (Diptera: Tephritidae), an important pest of fruit crops worldwide, is considered the most cosmopolitan and invasive of tephritid species (Malavasi 2009). In Brazil, C. capitata was reported for the first time in 1901, infesting citrus (Sapindales: Rutaceae) in the state of São Paulo (Ihering 1901). Despite the extensive territorial area of Brazil, C. capitata spread rapidly to other states, probably as a result of infested fruits being transported between localities. In the 1990s, the Mediterranean fruit fly reached the Brazilian Amazon (Ronchi-Teles & Silva 1996). To date, the only Brazilian states with no official report of the species are 4 states in the north (Acre, Amapá, Amazonas, and Roraima) and 1 in the northeast (Sergipe) (Silva et al. 2011b; Zucchi 2012).

Ceratitis capitata attacks a wide range of host plant species and has been reported on 88 host plant species in Brazil (Zucchi 2012). In the Brazilian Amazon, only 3 hosts of C. capitata are known, none of which are native to the region (Silva et al. 2011b). The species has been reported once in the state of Pará, in fruits of star fruit (Averrhoa carambola L.; Oxalidales: Oxalidaceae) and barbados cherry (Malpighia glabra L.; Malpighiales: Malpighiaceae), in the city of Belém (Silva et al. 1998). Although several fruit surveys were performed in an effort to find additional hosts, no new reports for the state were made until this report (Lemos et al. 2011; Zucchi 2012).

In 2012, 6 collections of fruits of Garcinia acuminata Planch. & Triana (sour bacuri) and 2 of Garcinia brasiliensis C. Martius (bakupari) (Malpighiales: Clusiaceae) were conducted in the experimental fruit orchard at Embrapa Eastern Amazon (1.466667°S, 48.453056°W), in the city of Belém, state of Pará, Brazil. At the collection site, we also recorded the presence of fruits on trees in orchards of A. carambola, M. glabra, Pouteria caimito (Ruiz & Pav.) Radlk. (Ericales: Sapotaceae), Eugenia stipitata McVaugh, E. uniflora L. (Myrtales: Myrtaceae), Psidium guajava L. and P. acutangulum D. C. (Myrtales: Myrtaceae).

Table 1.

Rates of infestation of Garcinia acuminata and Garcinia brasiliensis fruits by Ceratitis capitata. Belém, Pará, Brazil.

The samples were collected at random from plants with maturing or mature fruits. Fruits were removed from the plant or, if it had recently fallen, from the ground. Fruits were individually examined, and adult insects were obtained as described by Silva et al. (2011a). Selected specimens were deposited in the entomological collection at Embrapa Eastern Amazon.

We collected 508 fruits of G. acuminata and 150 of G. brasiliensis, both native to upland areas of the Amazon Forest (Lorenzi et al. 2006), totaling 9 kg of fruit (Table 1). Fewer G. brasiliensis fruits were collected primarily because this species is less abundant, but also in part because birds fed preferentially on G. brasiliensis instead of G. acuminata (pers. obs.).

In total, 361 puparia were obtained, from which 196 specimens of C. capitata emerged: 107 from G. acuminata and 89 from G. brasiliensis (Table 1). Although the area contained other fruit trees that could potentially be hosts of Anastrepha species (Diptera: Tephritidae), only C. capitata was obtained from the collected fruits.

Garcinia acuminata and G. brasiliensis are new hosts for C. capitata in the state of Pará. Garcinia brasiliensis is a previously reported host of C. capitata in the state of São Paulo (Souza-Filho 1999). The present study makes the first association between a fruit fly species and G. acuminata in Brazil.

The percentage of infested fruits was 3 times greater for G. brasiliensis (24.0%) than G. acuminata (8.0%). Garcinia acuminata showed the largest number of puparia per kg of infested fruits (363.7 ± 48.9) and the largest number of puparia per fruit (5.2 ± 0.6) (Table 1). Despite the small size of the fruits (approximately 4 cm long and 3 cm in diameter, average weight 10.2 g), a considerable infestation rate was observed, as well a pupal viability rate of nearly 50%.

The higher rates of infestation by C. capitata in G. brasiliensis than in G. acuminata may be explained by characteristics of the skin of its fruit. The skin of G. acuminata is tougher and may have influenced the death rate of last-instar larvae inside the fruits, as the tough skin may have prevented them from exiting and pupating. We propose this hypothesis after having observed dead larvae inside the fruit during the screening process. This was not observed for G. brasiliensis, whose fruits have a thinner and more flexible skin.

Considering the total number of fruits collected (not only infested fruits), the average infestation rates were low for both host species (30.7 puparia per kg for G. acuminata and 80.6 puparia per kg for G. brasiliensis), when compared with other hosts of C. capitata. In southeastern Bahia State, with C. capitata showing a predominance of 89.18%, 2 varieties of coffee (Coffea arabica L.; Gentianales: Rubiaceae) presented high rates of infestation, namely, 163.89 and 133.17 puparia per kg in shade-grown and sun-cultivated plantations of Catuaí Amarelo coffee, respectively, and 112.79 puparia per kg in suncultivated Novo Mundo coffee (Torres et al. 2010). Araújo et al. (2005) also reported high rates of infestation by C. capitata in star fruit (A. carambola), with 5.48 puparia per fruit and 118.8 puparia per kg of fruit, and in kumquat (Fortunella sp.; Sapindales: Rutaceae), with 2.16 puparia per fruit and 159.1 puparia per kg of fruit, in the municipalities of Mossoró and Assu, state of Rio Grande do Norte.

When fruits of G. acuminata were collected in 2008 in the same experimental orchard used in this study, no fruit flies were observed (Araújo et al. 2011). Infestation by C. capitata in the area is likely to be a recent event. The species has likely migrated from other hosts located near the experimental orchard, or from in natura fruit trading facilities (potentially infested fruits transported from various states in Brazil). Therefore, C. capitata may not be completely established at the study site, and it does not appear to be widely distributed in the region. Our report of G. acuminata and G. brasiliensis as new hosts for C. capitata in the state of Pará contributes towards a better understanding of the occurrence of this fruit fly in the Brazilian Amazon.

We thank the National Council for Scientific and Technological Development (CNPq) for the Research Productivity Fellowship granted to W. P. Lemos and R. Adaime.