A survey for parasitoids of the whiteflies, Bemisia tuberculata Bondar, Trialeurodes variabilis Quantaince, T. vaporariorum (Westwood), Aleurotrachelus socialis Bondar, Tetraleurodes sp., Aleuroglandulus malangae Russell and Aleurodicus sp., was conducted in 6 cassava growing regions of Colombia and Ecuador. In Colombia, the degree of infestation was predominantly high (>29 whiteflies/cm2) for A. socialis, B. tuberculata and T. variabilis in all cassava growing regions. In Ecuador, levels of infestations were high for Aleurodicus sp., A. socialis, B. tuberculata, Tetraleurodes sp. in the coastal region, and for T. vaporariorum in the Highlands. The parasitoid fauna of the whiteflies appeared to be more diverse in Colombia than in Ecuador. Eleven species of parasitoids representing 5 genera, 4 families and two superfamilies, as well as 1 hyperparasitoid, were collected from the cassava growing regions of Colombia and 4 species were collected from Ecuador. The parasitoids, Amitus macgowni Evans and Castillo, Encarsia sp., E. hispida De Santis, E. pergandiella Howard, E. bellottii Evans and Castillo, E. luteola group, E. sophia (Girault and Dodd), E. strenua group, Eretmocerus sp., Metaphycus sp. and Euderomphale sp., were collected. There were notable differences in parasitism among the different geographic regions and whitefly species. In general, Eretmocerus was the dominant genus in Colombia and Ecuador, followed by Encarsia sp. We found A. macgowni in regions characterized by high temperatures and bimodal rainfall. Percent parasitism per region surveyed ranged from 3 to 25% in Colombia and from 12 to 21% in Ecuador.
Whiteflies (Homoptera: Aleyrodidae) injure valuable agricultural commodities through mechanical feeding and virus transmission. Cassava, Manihot esculenta Crantz, is no exception to this rule, acting as a host of several species of whiteflies (i.e., Bemisia tuberculata Bondar, Trialeurodes variabilis Quantaince, Aleurotrachelus socialis Bondar, Tetraleurodes sp., and Aleuroglandulus malangae Russell (Castillo 1996)) in Colombia and of Bemisia tabaci (Gennadius) in Africa and Asia (Bellotti & Vargas 1986) where it vectors African cassava mosaic virus (ACMD). While this disease has not been reported yet in the Americas (Brown & Bird 1992), in Colombia, other diseases such as ‘cassava frog skin disease’ (CFSD) and common cassava mosaic are transmitted by B. tuberculata Bondar. Aleurotrachelus socialis is known to be the most important whitefly in the northern coast, eastern plains and western area of Colombia, but other species of whiteflies (e.g., T. variabilis, Tetraleurodes sp.) infesting cassava are poorly known (Castillo 1996).
Gold (1987) reported that cassava whiteflies in the area of Nataima, Tolima, Colombia are attacked by a complex group of natural enemies, including parasitoids, predators, and fungi and reported that among the natural enemies, parasitoids were more important mortality factors of cassava whiteflies than predators. Castillo (1996) and Evans and Castillo (1998) reported several cassava whitefly parasitoids in the northern cassava growing areas of Colombia. The parasitoids belong to the genera Encarsia, Eretmocerus (Hymenoptera: Aphelinidae) and Amitus (Hymenoptera: Platygastridae). Specifically, the species are Encarsia hispida De Santis, E. bellottii Evans and Castillo, and three undescribed species of Eretmocerus and Amitus macgowni Evans and Castillo.
The objectives of the present study were to determine the frequency of cassava whitefly parasitoid species in different geographical areas of Colombia and Ecuador.
Materials and Methods
This survey was conducted from April 1998 through June 2000 in the cassava growing regions of Colombia and Ecuador. The surveyed areas of Colombia were the Caribbean coast, Andean region, Valle Interandino del Cauca (Cauca), and Valle Interandino del Magdalena (Magdalena); the surveyed Ecuadorean regions were the coastal area and the highlands (Sierra). Geographic and climatic characteristics of each region are addressed in Table 1. In each area, the number of surveys ranged from 1 to 6 depending on cassava crop availability through the years. Each survey was conducted on 2-6-month-old cassava crops during periods of low or no rainfall in each surveyed area.
Sampling for whitefly species consisted of collecting a single leaf from the middle plant canopy from each of 100 randomly selected plants. A disc of 2.54 cm2 was excised from the leaf lobe that had the highest density of whitefly pupae and then placed in a 5-ml glass vial with 70% alcohol and transported to the laboratory. Whitefly density/ cm2 was grouped into three different categories: high (>29 pupae/cm2), medium (12-28 pupae/cm2) and low (<11 pupae/cm2). Whitefly pupae were identified with the keys of Caballero (1992; 1994) and Martin (1987). For further identification, pupae were sent to A. Hamon (Florida Department of Plant Industry and Consumer Services, Gainesville, FL).
To determine parasitism, 40 additional leaves were collected during each survey. Leaves were inspected for whitefly pupae, and the dominant whitefly species was identified. Once again, 2.54 cm2 of leaf were excised and those whitefly species with the lowest density in the sample were removed, leaving only the most abundant whitefly species in the sample. Samples were placed individually in 25-ml glass vials and held for 2-3 days at 24.5 ± 4°C and 70 ± 5% RH under laboratory conditions until parasitoids emerged. Emerging parasitoids were identified to genus with the taxonomic keys of Polaszek et al. (1992) for Amitus, Eretmocerus, Encarsia, Metaphycus and Signiphora; LaSalle & Schauff (1994) for Euderomphalini, and Rose & Zolnerowich (1997) for Eretmocerus. Each specimen was individually placed in a gel capsule vial and sent for further identification by G. A. Evans (Florida Department of Plant Industry and Consumer Services, Gainesville, FL) and M. Rose (Montana State University, Bozeman, MT).
Results and Discussion
Aleurotrachelus socialis, B. tuberculata, T. variabilis, and Tetraleurodes sp. were collected from the cassava growing regions of Colombia, confirming the results of Gold (1987), Arias (1995), and Castillo (1996). In Colombia, the degree of infestation was predominantly high (>29 whiteflies/cm2) for A. socialis, B. tuberculata, and T. variabilis in all cassava growing regions, with the exception of the Andean region, where T. variabilis was the dominant species (Table 2). The lowest degree of infestation (<11 whiteflies/cm2) was observed in the Caribbean coast for Aleurodicus sp. and A. malangae. We did not record T. vaporariorum from any of the surveyed cassava regions of Colombia (Table 2).
Levels of infestations were high for Aleurodicus sp., A. socialis, B. tuberculata, and Tetraleurodes sp. for the coastal region of Ecuador. We found T. vaporariorum, which commonly infests beans, Phaseolus vulgaris, for the first time at high infestation levels in cassava in the coastal and highlands regions of Ecuador (Table 2).
In general, A. socialis, B. tuberculata, Tetraleurodes sp., and Aleurodicus sp. were distributed in climatic regions characterized by high temperatures and extensive periods of drought (e.g., Caribbean, Magdalena). Trialeurodes sp. was found in higher numbers in mountainous regions, characterized by lower temperatures and high rainfall (Andean) (Table 1).
The parasitoid fauna of whiteflies appeared to be more diverse in Colombia than in Ecuador. Eleven species of parasitoids representing 5 genera, 4 families and two superfamilies, as well as 1 hyperparasitoid, were collected from the cassava growing regions of Colombia and 4 species were collected from Ecuador. Two of the Eretmocerus species are undescribed (Table 3) (M. Rose, pers. comm.). All other parasitoid species collected during this study were reported by Castillo (1996) and Evans & Castillo (1998). There were notable differences among the different geographic regions. On the Caribbean coast, A. socialis was parasitized by 8 species, with the genus Eretmocerus comprising 70% of the parasitoids (Table 4). In the Andean region, Eretmocerus sp., parasitized all whitefly species, but E. pergandiella was the predominant parasitoid of T. variabilis. The hyperparasitoid Signiphora aleyrodis Ashmead appeared in high densities in almost all sampled regions, probably reducing the efficacy of the parasitoids of A. socialis. In the Magdalena region, 73% of A. socialis were parasitized by A. macgowni, followed by Encarsia sp (26%). In the Caribbean, Andean, and Magdalena sampled regions of Colombia, B. tuberculata was parasitized by two undescribed species of Eretmocerus in addition to several described and undescribed Encarsia species. In the Cauca region the number of parasitoid species on A. socialis was almost the same as that collected on the Caribbean coast. However, the dominant genus was Encarsia (99%), represented by the species E. hispida, E. sophia, E. luteola, and E. bellotti.
The proportion of each parasitoid species collected from T. variabilis varied among the sampled geographical areas of Colombia. Encarsia was dominant in the Caribbean coast and in Andean region while Eretmocerus was dominant in Magdalena. In the Andean region, E. pergandiella was more frequent, followed by Eretmocerus sp., and E. hispida (Table 4).
The complex of parasitoids collected from whiteflies in cassava in Ecuador has only been identified to the generic level (G. Evans, M. Rose, pers. comm). In the coastal region, A. socialis was parasitized by Encarsia, followed by Amitus and Eretmocerus; Aleurodicus was parasitized by Euderomphale sp. Tetraleurodes and Trialeurodes were mostly parasitized by Eretmocerus sp., and B. tuberculata was parasitized by Encarsia sp., and Euderomphale. In the highlands region, T. vaporariorum was the only whitefly species collected, with approximately 16 pupae/2.54 cm2. The dominant parasitoids were Encarsia spp. representing 98% of the sample (Table 4).
In general, Amitus macgowni showed a localized distribution in those areas (e.g., Magdalena) with high temperatures and bimodal rainfall. With the exception of Ecuador, Eretmocerus was found both in warm regions and cool regions. Description and identification of the species within this genus will determine if the undescribed species are more frequent in some climatic areas than in others. Euderomphale was found in higher numbers in those areas (e.g., Coastal) with a low whitefly density, high temperatures, and minimal rainfall. E. pergandiella showed a general distribution among the different climatic regions (Magdalena, Andean), and particularly associated with Trialeurodes sp. The species E. hispida, E. sofia, E. bellottii, and Metaphycus sp., were found in the Magdalena region, characterized by high temperatures and a yearly average precipitation of 1,000 mm.
These observations indicate that both Colombia and Ecuador have a diverse parasitoid fauna attacking whiteflies on cassava. At the same time, these results indicate that the parasitism trend is influenced by the characteristics of each geographical area. During this study, some parasitoid species were discovered for the first time in some of the geographical areas of Colombia. In the Caribbean coast and in the Cauca region, A. socialis was parasitized by E. sophia; B. tuberculata was parasitized by E. sophia and Metaphycus sp. in the Caribbean coast, and by E. pergandiella and Euderomphale sp. in the Andean region. For the first time, E. pergandiella and E. sophia were collected as parasitoids of T. variabilis in the Caribbean coast, while E. hispida was the dominant parasitoid in the Andean region.
Because of the temporal and spatial limitation of our collections, parasitism of whiteflies on cassava will probably vary within the year, and data presented here may underestimate parasitism. For instance, high periods of parasitism may have been interspersed with periods of 0% parasitism. The highest frequency of parasitoids was obtained in A. socialis, which in general had the highest density in most of the surveyed regions (Table 5). Low levels of parasitism were not uncommon and ranged from 3-5% in the Andean and Cauca regions of Colombia, 10-12% for the Caribbean and Highlands regions of Colombia and Ecuador, respectively, to 21 and 25% in the coastal region of Ecuador and Magdalena region of Colombia, respectively (Table 5). These data suggest that parasitoids are ineffective in reducing cassava whitefly populations in the surveyed areas of Colombia and Ecuador. However, it is necessary to do a more thorough study on those parasitoids that cause higher mortality. For instance, A. macgowni was observed as the dominant parasitoid of A. socials in the Magdalena region. Therefore, studies toward augmentation and conservation of A. macgowni should be encouraged in that region. Life cycle and behavioral studies of Encarsia pergandiella as an important parasitoid of Trialeurodes sp., should be conducted.
During this study, whitefly densities were low in Ecuador. Therefore, further studies are necessary to properly determine the potential of each parasitoid species in that country.
We thank R. Cave for suggestions to improve the manuscript. We thank G. Evans, M. Rose, and A. Hamon for insect identification. This work was supported by a grant from USAID to the International Center for Tropical Agriculture (CIAT). Florida Agricultural Experiment Station Journal Series No. R-09645.
Climatic and geographical range of Cassava growing regions included during the 1998-2000 survey in Colombia and Ecuador.
Number of Whiteflies recorded on Cassava leaves in 6 geographical areas of Colombia and Ecuador.
Parasitoids emerging from Cassava whiteflies collected in Colombia and Ecuador.
frequency of parasitoids from whitefly species in 6 geographical regions of colombia and ecuador
Number of emerging parasitoids from Cassava whiteflies in 6 geographical areas of Colombia and Ecuador.