Bemisia tabaci (Gennadius) is a cryptic whitefly species, owing to lack of morphological characters that permit recognition of behavioral and/or genetic variants (Mound 1963; Mohanty & Basu 1986; Gill 1992; Gawel & Bartlett 1993; Bedford et al. 1994) and, consequently, has been considered a species complex (Brown et al. 1995; Rosell et al. 1997; Frohlich et al. 1999) or a complex species (Perring et al. 1993; Bellows et al. 1994). The B. tabaci complex is abundant in cultivated and uncultivated subtropical and tropical habitats worldwide and is unusual among whiteflies in that it is polyphagous on herbaceous plant species, whereas most whitefly species are monophagous on woody species (Mound 1963; Lopez-Avila 1986; Martin 1987). Among the few whitefly species that colonize herbaceous hosts, B. tabaci is unique in that its host range is highly variable across the species. Host range phenotypes vary from restricted to highly polyphagous, illustrating the key character that led to the recognition of ‘host races’ for B. tabaci (Bird 1957).
Host races are one variant of biotype (Diehl & Bush 1984) of B. tabaci, now referred to as ‘biological types’, that exhibit variation in geographical distribution, host range, fecundity, dispersal behavior, insecticide resistance, natural enemy complexes, and endosymbiont complement (Rowland et al. 1991; Costa et al. 1993a; Costa et al. 1993b; Bedford et al. 1994; Costa et al. 1995; Kirk et al. 2000). Some of these factors may influence biotype evolution and possibly the formation of ‘new’ biotypes.
The first recorded whitefly outbreak occurred on tobacco in Greece in 1889 and led to the description of the whitefly as B. tabaci (Genn.) (Gennadius 1889). This insect was reported in Florida within the same decade (Mound 1963; Hamon & Salguero 1987), but was not considered a pest in the state until 1986 when large populations infested poinsettia (Price 1987). The appearance of a previously undescribed tomato irregular ripening disorder and a squash silverleaf disorder were associated with the introduction of this exotic biotype in Florida (Schuster et al. 1990; Schuster et al. 1991). The spread of geminiviruses in beans and tomatoes (Blair et al. 1995; Polston & Anderson 1997) was also associated with the introduction of this new B. tabaci biotype. The new biotype (a.k.a. B. argentifolii Bellows & Perring, the silverleaf whitefly) was designated B to separate it from the original A biotype. The objective of this work was to conduct an extensive survey of B. tabaci populations in Florida agricultural ecosystems to determine if the B biotype had excluded non-B biotypes.
During the 2000-2001 growing seasons, an extensive survey was conducted in 13 locations, representing 8 different counties and corresponding principal vegetable producing areas in Florida (FDACS/FASS 1998), including 15 economically important agricultural crops and eight weed hosts found in proximity to the crop fields (Table 1). The same crops were surveyed across locations, when possible, and many counties were sampled multiple times. Adult whiteflies were collected at each location by vacuuming host plants with a Makita Cordless Cleaner (Model 4073D; Anjo, Aichi, Japan) outfitted with size 12-dram plastic collection vials (BioQuip Products, Gardena, CA). The plastic bottom of the collection vials was cut out, screened and placed directly into the hole made for vacuum attachments for easy sampling of host plants. At each location, leaves from crop and weed plants were collected to obtain whitefly nymphs. Adult whiteflies were counted and sexed, nymphs were removed from host leaves, and all whitefly samples were stored in 95% ethanol for molecular analysis. At the International Center for Tropical Agriculture (CIAT), in Cali, Colombia, whitefly species were verified as B. tabaci by classical morphological characteristics of the 4th nymphal instars and adult whiteflies were biotyped by RAPD/PCR analysis following methods adapted from De Barro and Driver (1997). CIAT has maintained whitefly colonies of the A biotype on multiple hosts since the mid-1980s. The B biotype of B. tabaci was detected and characterized in Colombia by Quintero et al. (1998), and colonies of this biotype have been maintained at CIAT since that time. Thus, CIAT provided the positive controls for this study. Whitefly data (mean number of nymphs, male, female and sex ratios (female: male) were analyzed by the General Linear Models (GLM) procedure where hosts were sampled more than once, and differences among hosts were determined by Ryan-Einot-Gabriel-Welsch multiple-range test (REGWQ) at α = 0.05 (SAS Institute 2000).
A total of 9,963 nymph and adult B. tabaci were collected in 13 locations across Florida (nymph = 3,364; female = 5,061; male = 1,538). Sex ratios (female: male) were not different among host plants (F = 2.23; df = 9,21; P = <0.0629) and ranged from 0.63 on pea to 0.91 on Cucurbita spp. (Table 2).
RADP/PCR analysis of statewide whitefly samples using Primer set H16 (Fig. 1) indicated that only the B biotype of B. tabaci was present. Gels for all locations and hosts were identical to Fig. 1 (data not shown). Primer set H9 was used (data not shown) to confirm the biotype B origin of each whitefly sample. Band pattern variation in lanes 14 (Fig. 1a) and 12 (Fig. 1b) represent natural within biotype variation related to the RAPD method used as described by De Barro and Driver (1997). Minor band variability within the 600 to 900 bp region demonstrates some polymorphism between different B biotype individuals. However, the real area of interest is the region between 300 and 600 bp which demonstrates a consistently reproducible pattern (520, 500, and 344 bp) that is unique to the B biotype (De Barro and Driver 1998). These bands are not present in the case of other biotypes studied.
The authors thank Kendall Farms, Boynton Beach; David Neil, Neil’s U-Pick, Fort Pierce; Tommy Whitworth, Whitworth Farms, Boynton Beach; Tom Young, Rogers Novartis Seeds, Inc., Naples; Ted Winsberg, Green Cay Farms, Boynton Beach; Jack Martin, Jack Martin Farms, Boynton Beach; Paul King, Uniroyal Farms, Sanford; and Kenneth Hall, Suwannee Farms, O’Brien, for allowing us to survey their crops. We also thank Dr. Joe Funderburk (University of Florida, North Florida Research and Education Center, Quincy), Dr. Dave Schuster (University of Florida, Gulf Coast Research Station, Bradenton), and Dr. Bob McMillan Jr. (University of Florida, Tropical Research and Education Center, Homestead) for assistance in surveying their respective research facilities.
Summary
An extensive survey of B. tabaci populations in 15 economically important crops and 8 weed species in Florida was carried out from March 2000 through May 2001. Sex ratios did not significantly differ among host plants. Biotype analysis by RADP/PCR indicated the presence of only the B biotype of B. tabaci in all collections. These data suggest that in Florida the B biotype of B. tabaci has excluded the native non-B biotypes in agricultural ecosystems.
Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U.S. Department of Agriculture and does not imply its approval to the exclusion of other products that may also be suitable.
References Cited
Fig. 1.
Electrophoretic profiles of amplified DNA obtained for Bemisia tabaci samples collected from agricultural ecosystems in Florida with RAPD-PCR analysis. L, molecular weight marker Gibco-BRL 1-Kb; A, + control for Bemisia tabaci biotype A; B, + control for Bemisia tabaci biotype B; C, (-) water control with H16 primers. Gel 1a, lanes 1-10 sampled from common bean, Phaseolus vulgaris, Kendall Farms, Homestead, FL; Gel 1a Lanes 11-16 and gel 1b lanes 1-4, sampled from tomato, Lycospersicon esculentum Mill., U. S. Horticultural Research Laboratory, Fort Pierce, FL; Gel 1b, lanes 5-14 sampled from tomato, L. esculentum Mill., Neil’s U-Pick Farm, Fort Pierce, FL.
