Extraction Methods--Rates of Disappearance There has been some concern over the method of pesticide ex- traction, particularly for harvester exposure residues. For ap- plicators and mixer-loaders, methods can be developed as needed with defined substrates. For the extraction of leaf, fruit, and soil surface residues, peculiar to harvester exposure studies, a stan- dard methodology has been adopted by many researchers (Iwata et al., 1977; Spencer et al., 1977). Fruit and leaf surface residues are recovered with organic solvents from a mild soap solution in which they have been shaken. Soil surface residues are recovered by vacuuming surface soil through a 100-mesh screen. However, at least for foliar residues, some experimenters shake leaves in organic solvents (Ware et al., 1975, 1980). These organic solvent residue data may be higher and lead to slower calculated rates of disappearance, making it appear that the worker is exposed to higher residues of longer duration. Models of exposure based on the soap solution method have been and are being produced. A model developed for one chemical is then used for another. Sol- vent residue data for a chemical could be alternatively used in these models once the relationship between the organic solvent and soap solution methods is understood and quantified. Harvester Exposure Harvester exposure to pesticides has been the subject of several reviews (Davis, 1980; Gunther et al, 1977; Nigg and Stamper, 1982). Since pesticide may be transported to the harvester primarily on surface dust, the dermal exposure pads are faced with 16-ply surgical gauze. Respiratory exposure is usually not measured in harvester experiments. Using the total body exposure estimation method, with dermal pads and handwashes, two models of tree harvester exposure as a function of leaf residue level have been produced (Popendorf and Leffingwell, 1982; Nigg et al., 1984). These are substantially the same model and agree with unpublished data from Washington apples (Davis, J., personal communication). Applicator exposures may range from 69 mg/h (Wojeck et al., 1982) to 15,000 mg/h (Wojeck et al., 1981) while tree fruit harvester exposures range from 0.07 mg/h (Spear et al, 1977) to 2.35 mg/h (Nigg et al., 1984). While harvesters are exposed to less contaminating material per se than applicators and mixer- loaders, the quality of their exposure may be different. The ap- plicator or mixer-loader is exposed to the parent compound only, whereas the harvester or any laborer reentering a treated area may be additionally exposed to a metabolite many times more toxic than the parent. Is one situation more dangerous than the other? Acute poison- ing cases for both situations are documented. We return to this point later in connection with protective strategies. It would ap- pear that the chronic liability to an applicator/mixer-loader would be potentially greater because of the larger dose. Signifi- cant exposure to the small acreage farmer who participates in all farming operations is probably both chronic and acute. Application Methods Pesticide exposure rates measured during various application methods are presented in Table 1. Unfortunately, most of these studies employed different experimental designs. Either the der- mal exposure pads were located differently or the estimated total body doses were made solely on the basis of body areas not covered by normal work clothing. In some cases (Wojeck et al., 1981, 1982, 1983) calculated doses were determined as if no clothing were worn at all. Nonetheless, a rough order-of- magnitude comparison is justified. It shows that the airblast method generally leads to more exposure, a point also made by Wolfe et al. (1972). Boom sprayer exposures are higher than for handsprayers, and handsprayers are more exposed than helicopter loaders. 230 TABLE 1. Estimated total body exposure for various pesticide application methods (mg/h). Application method Pesticide Mean exposure level Reference Airblast Carbaryl 59.1 Comer et al., 1975 Airblast Ethion 1850 Wojeck et al., 1981 Airblast Lead arsenate 95 Wojeck et al., 1982 Airblast Various pesticides 200 Wolfe et al., 1972 Helicopter-Loader 2,4-D .63 Lavy et al., 1982 Moderate Boom Paraquat 168.6 Wojeck et al., 1983 High Boom Paraquat 18.4 Wojeck et al., 1983 Low Boom Paraquat .40 Staiff et al., 1975 Low Boom (Shielded) Paraquat 28.5 Wojeck et al., 1983 Low Boom DNOSBP 131.1 Wolfe et al., 1961 Low Boom Na-DNOC 39.6 Wolfe et al., 1961 Low Boom Diallate 70 Dubelman et al., 1982 Handgun-Airboat Diquat 1.82 Wojeck et al., 1983 Hand Sprayer Paraquat .29 Staiff et al., 1975 Handgun Carbaryl 55 Leavitt et al., 1982 Body Areas The studies in Table 1 generally agree on two counts. Hands account for 60-95% of the total estimated exposure for ap- plicators and mixer/loaders, and in almost every case metabolites can be found in the urine. Although a comparable quantity of data is not yet available for harvesters, gloves may be a useful protective device (Wicker et al., 1979). Hand exposure is higher than forearm exposure for strawberry harvesters (Zweig et al., 1983). However, in a study where pads were placed on various body areas of citrus harvesters, hands accounted for only about 10% of the total estimated ex- posure (Nigg et al., 1984). Protective Strategies The major requirement for protecting farm workers from pesticide exposure is reliable information. Many chemical com- panies market pesticides by touting the low toxicity of the pesticide formulation. The formulation may affect the dose, but not the intrinsic toxicity of the chemical. A good general rule is that the more acutely toxic the chemical, the more pesticide poisoning cases it will produce. The pesticide salespeople, however, may not even know the toxicity of the active ingredient. They are exposed to little, if any, of their product so that safety is an easy claim to make. In our experience, those salespeople who have been poisoned take a reticent approach. The poisoning history of the pesticide is of critical importance and, while chemical companies may have this information, they certainly do not advertise it. The argument that use levels also contribute to poisoning cases is logical, but this is a misleading and dangerous argument. Who can predict the actual use level of a chemical? Toxicity alone is the most important factor. There are several basic precautionary measures suggested by the data in our cited references. 1. Regardless of the application method, the hands of ap- plicators and mixer-loaders receive the highest level of ex- posure. Frequent washing of the hands with soap and water, and the washing of equipment prior to mainten- ance, appear to afford the best protection. Cloth gloves are PROCEEDINGS of the CARIBBEAN FOOD CROPS SOCIETY-VOL. XX