Analitic Methods

Analitic Methods
Field and greenhouse studies which report concentrations of arsenic, cadmium, copper, lead,
mercury, nickel, selenium, or zinc in both surface soil and collocated, aboveground plant tissue were identified. Most plant species were agricultural crop plants. For some elements, many studies were pot studies in which inorganic salts were added to soil. Information regarding soil and plant concentrations, soil parameters, exposure time, chemical form, dry or wet weight, extraction method, plant species, and plant part was compiled in a spreadsheet. Only studies in which concentrations were expressed on a dry weight basis were used. Some soils were air dried rather than oven dried. Although most studies reported that plant material was washed, studies were not excluded if the extent of washing was not stated in the paper. Studies were used even if the individual investigators observed no correlation between concentrations of contaminants in soils and plants (e.g., arsenic in Norway spruce, Wyttenbach et al. 1997; copper in Sitka-spruce seedlings, Burton et al. 1984; copper in radish foliage, Davies 1992). Concentrations of chemicals in soil or plants were sometimes estimated visually from a figure, but only
if estimates could be made within about 10%. Studies were not used if the only plants tested were those known to hyperaccumulate elements.
Each plant species or variety, soil type, location, concentration of the test element in soil, and form of an added element represented an independent observation in the dataset. Differences in exposure duration or above-ground plant part did not constitute separate observations. That is, concentrations in soils or plants that differed on the basis of one of these two variables were averaged. (The number of observations in these means, which ranged between 1 and 6, was not retained in the subsequent statistical analysis.) For example, concentrations of nickel in upper and lower leaves of bush bean (Sajwan et al. 1996) and concentrations of lead in corn leaves and stalks (de Pieri et al. 1997) were averaged and each constituted a single observation. Also, concentrations of lead in spruce needles (Nilsson 1972) and cadmium in clippings of red fescue (Carlson and Rolfe 1979) after different periods of exposure were averaged. A pattern of higher levels of accumulation with increased exposure time was not generally observed. The database of bioaccumulation concentrations is presented in Appendix A.
Concentrations of contaminants in soil at the time of plant sampling were used if known. If these
concentrations were not measured (as was often the case in pot studies), the initial concentration of the element measured in or added to soil was assumed to be equivalent to the final oncentration. In field experiments, the change in soil concentration of an element over time was assumed to be minimal (e.g., selenium in van Mantgem et al. 1996). However, total soil concentrations of elements in pot studies have been observed to change as much as twenty percent during an experiment. The concentration of an element in soil prior to the addition of the salt in a pot study was often not stated. Thus, the added concentration was often assumed to be equivalent to the total concentration. Experimental treatments or field studies in which aerial contaminants potentially contributed to uptake were excluded from the database. In some early field studies with lead, aerial exposure to lead additives from gasoline was likely (e.g., Parker et al. 1978). In other field studies, ongoing exposure to metal contaminants from smelters or other sources was possible, though data from the vicinity of a smelter or other air source were not used unless it was demonstrated in the study that air was not a significant route of contamination.
Observations were included in the database if the total chemical concentration in soil was measured, either by extraction with strong acid or by extraction with moderately strong acid (e.g., 4 N sulfuric acid) sometimes accompanied by heat. In one study, it was shown that extraction of arsenic with 6M HCl for 2 h under constant rotation gave the same recovery as digestion in aqua regia, a mixture of concentrated nitric and hydrochloric acids (Otte et al. 1990). Studies in which concentrations of contaminants in soil were determined by a partial extraction with DTPA (diethylene triamine pentaacetic acid), weak acids or water were excluded from analysis, unless DTPA was used only to extract the background fraction of the element, and salts were added. Although concentrations of DTPA-extracted contaminants from soils sometimes correlate with those taken up by plants (Sadiq 1985), this estimate of bioavailability has been observed not to be valid for some metals (Sadiq 1985, Sadiq 1986, Hooda and Alloway 1993) or for comparisons of soils of varying pH.