L. Polette1, J.L. Gardea-Torresdey1, R.R. Chianelli1, I.J. Pickering2, and G.N. George2

1Department of Chemistry, The University of Texas at El Paso, El Paso, TX 79968, Phone: 915-747-5359, FAX: 915-747-5748; 2Stanford Synchrotron Radiation Laboratory, Bldg. 120, Menlo Park, CA 94025, Phone: 415-926-4604, FAX: 415-926-4100


Metal contamination in soils has become a widespread problem. Emerging technologies, such as phytoremediation, may offer low cost cleanup methods. We have identified a desert plant, Larrea tridentata (creosote bush), which naturally grows and uptakes copper and lead from a contaminated area near a smelting operation. We determined, through chemical modification of carboxyl groups with methanol, that these functional groups may be responsible for a portion of copper(II) binding. In contrast, lead binding was minimally affected by modification of carboxyl groups. X-ray absorption spectroscopic studies conducted at Stanford Synchrotron Radiation Laboratory (SSRL) further support copper binding to oxygen-coordinated ligands and also imply that the binding is not solely due to phytochelatins. The EXAFS data indicate the presence of both Cu-O and Cu-S back scatters, no short Cu-Cu interactions, but with significant Cu-Cu back scattering at 3.7Å (unlike phytochelatins with predominantly Cu-S coordination and short Cu-Cu interactions at 2.7Å). Cu EXAFS of roots and leaves also vary depending on the level of heavy metal contamination in the environment from which the various creosote samples were obtained. In contrast, Pb XANES data of roots and leaves of creosote collected from different contaminated sites indicate no difference in valence states or ligand coordination.

Keywords: Larrea tridentata, phytoremediation, creosote bush, x-ray absorption


Soils are generally a significant source of nutrients for plants. The nutrients provided by the soil can be absorbed onto the cells of the plant root, then translocated into various other tissues of the plant. The availability of nutrients in soil is usually dependent on lithogenic and pedogenic factors; however, elevated levels of elements in soil such as heavy metals, are usually the result of anthropogenic activity (Peterson & Girling, 1981; Narwal, Singh, & Singh, 1991; Fala-Ardakan, 1984; Thomas, Rohling, & Simon, 1984).

Similar mechanisms that allow nutrients to be uptaken in plants may also allow heavy metals present in the soil, which are generally indistinguishable from nutrients, to also be uptaken. Trace levels of metals such as copper and zinc are vital to such plant systems because of their participation in oxidation, electron transfer, and various enzymatic reactions, but high quantities of heavy metals such as copper and lead can be physiologically injurious (Hamer, Thiele, & Lemontt, 1985; Fernandes & Henriques,1991; Slivinskaya, 1991; Sastry & Chaudhary, 1989; Vojtechova & Leblova, 1991; Kastori, Petrovic, & Petrovic, 1992). The ability of plants to sequester high levels of heavy metals, however, is not always deleterious. Certain plant species naturally grow near industrially polluted vicinities (Motto, et al., 1970; Nriagu & Pacyna, 1988; Kansanen & Venetvaara, 1991; Worthington, 1989; Bache, et al., 1991; Narwal, Singh, & Singh, 1991). Consequently, many researchers are investigating the opportunity to use plants to alleviate metal contamination in soils. This technique is known as phytoremediation, and holds the promise of being both a feasible and cost-effective process of reclamation and remediation of contaminated soils.

Larrea tridentata, commonly referred to as the creosote bush, is the most common desert shrub of the Southwest, covering roughly 20 million acres from western Texas to California. The creosote bushes are found naturally growing in the heavy metal contaminated soils near a copper smelting operation in El Paso, Texas. The industrial activity in the area has resulted in the accumulation of many heavy metals in the soil including copper, lead, cadmium, and zinc. To date, few studies have been performed demonstrating creosote bushes as metal scavengers. Because creosote bushes are so prevalent throughout the Southwest and because they are able to grow in such contaminated soil environments, they might possess the qualities necessary for utilization as phytoremediation resources, especially in arid environments. Most of the conventional methods for soil cleanup are very expensive and only marginally effective. Furthermore, the technologies typically employed to decontaminate the soils can increase contaminant exposure to cleanup crews and to nearby residents. Because our data demonstrate the ability of creosote bushes to sequester heavy metals from the soils, phytoremediation of contaminated soils using creosote bushes may have an enormous economic value. Our data already suggest that creosote bushes are able to uptake heavy metals from soil, but to increase the success of creosote as well as other plant species in commercial-scale phytoremediation, it is important to understand the biological mechanisms by which plants are able to accumulate metals and survive. As previously described, the initial process of plant interactions with nutrients or heavy metals occurs with the cell walls or membranes. Specifically, these exterior surfaces have a common composition of proteins and carbohydrates, including carboxyl groups, with which metallic ions such as Cu(II) could react; such bond formations could be accompanied by displacement of protons dependent in part on the extent of protonation as determined by pH (Crist, et al., 1981; Gardea-Torresdey, et al., 1990). Part of the research here describes the role of carboxyl functional groups in the binding of Cu(II) ions. Additionally, because heavy metals in plant systems can bind to a variety of ligands and in different valence states, there are a limited number of techniques which can provide complete structural information and binding mechanisms. However, through the use of X-ray absorption near edge structure (XANES), extended X-ray absorption fine structure (EXAFS), and chemical modification studies herein, we elucidate initial models that describe local structural information and interatomic distances of copper and lead binding in the various tissues (i.e., roots, leaves, and stems) of creosote bush.


Collection of Larrea tridentata

Triplicate samples of creosoted were collected from various measured distances from a smelting operation and then dried in an oven at 900C; they were then separated into roots, leaves, and stems. The tissues were washed with deionized water and 0.01M HCl and were then acid digested according to EPA method 3050 (United States EPA, 1987) and method 200.3 (United States EPA, 1991). Analysis of the metals of interest were performed using a Perkin Elmer model 3110 Atomic Absorption Spectrophotometer (AAS) with deuterium background subtraction. Impact bead was utilized to improve sensitivity. Samples were read three times and a mean value and relative standard deviation were computed. Calibrations were performed in the range of analysis and a correlation coefficient of 0.98 or greater was obtained.

EXAFS and XANES Analysis

Samples were obtained as described above. Additionally, however, to see if binding sites varied depending on sample preparation, fresh leaves (not dried or ground) were obtained from the field, washed, and sealed; and XANES and EXAFS data were obtained for copper and lead within 42 hours of collection. XAS data were collected on beamline SB07-3 (1.8 T Wiggler field) using a 13-element germanium array detector at the Stanford Synchrotron Radiation Laboratory. During data collection, samples were maintained at a temperature of approximately 10K using an Oxford Instruments CF1204 liquid helium flow cryostat. Copper K-edge and lead LIII- edges were used to probe the chemical forms of the metals. While scans were accumulated, the absorption of a metal foil was measured simultaneously by transmittance.

Chemical Modification Study

Although two of the sites (site 2 and site 4) where creosote was collected contained high concentrations of copper and lead, site 8 creosote samples were collected as controls (i.e., no metals in the plant tissues or soils). For chemical modification studies then, only site 8 creosote biomasses were used. Chemical modification of creosote carboxyl groups with acidic methanol was carried out in accordance with a previously described procedure (Darnall and Gardea-Torresdey, 1989) according to the following reaction:

In summary, a 4.5 g sample of roots of creosote biomass was added to a 600ml beaker followed by 316.5 ml of 99.9% trace metal grade methanol and 2.7 ml of concentrated hydrochloric acid to give a final solution concentration of 0.1 M. The same procedure was repeated for control samples with the exception that 316.5 ml of methanol was replaced with 316.5 ml of DI water. The esterification reaction was performed at room temperature. After 6, 12, and 24 hrs, 25 ml of each sample was removed and the reaction terminated by 1) centrifuging the sample at 3000 RPM for 10 min, 2) removing the acidic supernatant, and 3) washing the sample three times with deionized (DI) water. After 48 hrs, the remaining esterified biomass solutions were also quenched accordingly. The above mentioned procedure was repeated for the leaves. Cu(II) and Pb(II) binding experiments to esterified and unesterified samples were then done at pH 2 and pH 5 using 6.3 ppm (0.1mM) solutions of either copper (as CuSO4) or lead as Pb(NO3)2 according to a previously described procedure (Gardea-Torresdey, et al., 1996a). The samples were then centrifuged and the supernatants were analyzed for copper and lead content using AAS.

After determining the extent of Cu(II) and Pb(II) binding to the esterified and unesterified creosote biomasses, the metals were removed from the pellets using 0.1 M HCl. Following removal of the metals from the pellets, the samples were treated with 2 ml of 0.1 M NaOH in order to hydrolyze the esterified carboxyl groups. After one hour of shaking, the reactions were terminated by centrifugation followed by removal of the basic supernatant and washing of the pellet three times with DI water. Again, the Cu(II) and Pb(II) binding abilities of the hydrolyzed creosote biomass were determined at pH 2 and pH 5. The percent metal adsorbed or bound was obtained by calculating the difference between the initial metal concentration applied to the biomass and the final metal concentration.


Through acid digestion followed by analyses by atomic absorption spectroscopy, concentrations of lead in roots, leaves, and stems were found (Gardea-Torresdey, et al., 1996b). Specifically, out of the eight sites sampled, the highest levels of heavy metals of copper and lead were observed in site 4, followed by site 2, while site 8 samples were found to contain no copper or lead; thus site 8 data exemplify control samples. Site 4 lead data for roots, leaves, and stems were: 650, 150, and 110 mg/Kg, respectively, while copper concentrations were 953, 493, and 370 mg/Kg, respectively. Concentrations of copper and lead were also high in site 2, but lower than those of site 4, as previously reported (Gardea-Torresdey, et al., 1996b). In order then to probe the chemical environment of the copper and lead species, EXAFS and XANES data were collected. XANES analyses are based on the fact that X-ray energies sufficient to eject core electrons from specific elements of choice result in an increase in absorption. The absorption edge can provide insight into the valence states of the element of choice, copper in this case. Figure 1 shows the copper K-edge XANES of L. tridentata leaves and roots from two different sites located near the smelter operation. While the spectra from the different sites are similar, there are significant differences between the spectra of the leaves and roots, indicating that the chemical forms of copper differ between leaf and root tissue. The spectra from Figure 1 were compared with model compound data (Pickering, et al., 1993). Spectra are consistent with the presence of both cuprous and cupric copper in both the roots and leaves, with a larger fraction of cuprous copper in the leaf tissue.

EXAFS data result from sinusoidal oscillations as the electrons travel in space and interfere with the wave functions of near-shell neighboring atoms. EXAFS analyses can provide interatomic distances, identify near-neighboring atoms, and in some cases can provide approximate coordination numbers. Figure 2 shows a comparison of the copper K-edge EXAFS Fourier transforms of fresh L. tridentata leaves, and of a copper phytochelatin sample (isolated from a fission yeast) which is thought to be involved in a low-metal detoxification in higher plants (Winge, et al., 1993). The copper phytochelatin data are typical of the metallothionein class of cuprous-thiolate binding proteins (Pickering, et al., 1993; Winge, et al., 1993), with predominantly Cu-S coordination and short Cu-Cu interactions at 2.7Å. The copper EXAFS data indicates the presence of both Cu-O and Cu-S back scatters (unlike the phytochelatin with only Cu-S), no short Cu-Cu interaction, but with a significant Cu-Cu back scattering at 3.7Å. Thus, this preliminary evidence suggests that the chemistry of copper binding in creosote bushes may be significantly different than that suggested by the established dogma. It is important to note here, however, that the metal binding to metallothionein type polypeptides can be broken down into three classes: I, II, and III. Phytochelatins generally fall into the category of class III, in that the polypeptide chains do not follow the Cys-X-Cys sequence where X is an amino acid other than Cys; rather, the class III polypeptide sequence is g-Glu-Cys-Gly and the peptides are nontranslationally synthesized (Rauser, 1990). Regardless, however, of the variations in the three classes of metallothionein type metal binding, the Cu-Cu interactions at 3.7Å in creosote are atypical of any class.

Figure 3 shows the transform magnitudes for inactivated site 4 leaf tissue, fresh leaf tissue from site 4, and inactivated site 2 leaf tissue. In all cases, there is a Cu-O peak located at 1.85Å. The shoulder peaks located at approximately 3.2Å and 3.9Å for site 2 leaves are clearly different than the Cu-Cu interaction peaks at 3.7Å for site 4 leaf tissue. Such pronounced transform peaks are generally not observed in this region, except in the case of multiple scattering. The peaks between 3-4Å then, could be due to multiple scattering, which is indicative of a linear bridged copper complex. The Cu-S peaks at 2.2Å, as well as the small Cu-Cu interaction peaks at 2.7Å could indeed be attributed to the multimetallic copper-thiolate clusters as proposed in typical phytochelatins (Pickering, et al., 1993). However, the Cu-O and large Cu-Cu interaction is a novel copper binding mechanism which has not been previously reported in the literature. Thus, we speculate that creosote possesses two distinct mechanisms for binding copper, a phytochelatin and a linear Cu-O-Cu interaction. Theoretically, in a linear coordination bonding scheme, such as Cu-O-Cu, the two bridging bond distances can be added. Specifically, for site 4, the Cu-O bond distance of 1.85Å, when multiplied by two, gives the exact Cu-Cu distance of 3.7Å as depicted. The two Cu-Cu interaction peaks for site 2 though, are not as readily accounted for. However, if one considers the atomic radii differences between a Cu(I) and a Cu(II) species, it could be postulated that the site 4 leaf tissue contains a linear Cu(I)-O-Cu(II) bond while the site 2 leaf tissue contains a mixture of a linear Cu(I)-O-Cu(I) and Cu(II)-O-Cu(II) species. These data suggest that the biochemical mechanisms responsible for uptake and binding are different depending on the plant growth conditions.

Based on numerous examples from the literature, as well as the data presented here, the Cu(I) ions can be attributed to ligation to cysteinyl sulfurs; but the nature of the Cu(II) and Cu-O species are not yet completely understood. As previously mentioned though, a variety of chemical functional groups such as carboxylic acids reside on the cell walls of plants. It has been suggested that these functional groups, under certain pH conditions, could be available for the binding of metallic ions such as Cu(II) (Crist, et al., 1981; Gardea-Torresdey, et al., 1990). Also, the Fourier transform data indicate that there is no evident difference between leaves from site 4 which have been oven dried and ground and fresh leaves. Based on this information, it was believed that the functional metal binding groups are stable with respect to various sample treatments. Therefore, we attempted to determine the role of Cu(II) binding in creosote bush to carboxylate groups. Through esterification reactions, we found that copper binding at pH 5 was decreased by as much as 33.9% in the roots of creosote and as much as 11.2% in the leaf tissue (data not shown). At pH 2, there was little to no binding to either the esterified or the control samples, which was expected considering that the high concentration of hydrogen ions would compete with the metal carboxylate binding sites. Also, Pb(II) binding to esterified or unesterified samples (controls) of roots or leaves was minimally affected at either pH 2 or pH 5 (data not shown). This may suggest that the binding sites for Cu(II) and Pb(II) in creosote leaves and roots are different.

Figure 4 edge data shows how Cu binding to inactivated site 8 control leaves differs from copper binding in creosote grown naturally in contaminated soils. The first features in the XANES correspond to transitions to the lowest unfilled electronic levels of the absorbing metal element (Lytle, Lytle, & Smith, 1996.). Specifically, the edge peak evident in site 4 is due to Cu(I) while the two larger peaks at ~8990 eV in both leaf samples from sites 2 and 4 are indicative of Cu(II), as the shift in energy corresponds to an increase in valence state.

Figure 5 shows the Cu-K edge EXAFS Fourier transforms for the control site 8 leaves which were reacted with Cu(II), as well as comparisons to site 4 leaf and root tissue. The Cu-O peak for site 8 leaves has shifted by approximately 0.1Å and more closely resembles the transform data for the roots. The shoulder peak for site 8 at 2.2Å is due to Cu-S interaction and suggests that the Cu binding still may partly occur through sulfur ligation. More importantly though, the differences in the spectra of Figure 5 further support the fact that there is not an apo site, but rather, the metal binding sites could be enzymatically produced since the Cu-S interactions were not significantly observed on the control leave tissues which were reacted with copper.

Figure 6 shows Pb L III edge data of roots, fresh leaves, and site 8 (control) leaves reacted with Pb. Lead EXAFS data are difficult to interpret and thus a dearth of information exists on Pb, owing to the fact that lead possesses specific electronic and structural disorder, including static disorder and thermal vibrations which can smear out EXAFS oscillations (Manceau, 1996). As a result, the lead XANES spectra below is only preliminary, while further lead data will be examined during subsequent beamtime at SSRL. The data do show, however, that there are no differences in valence states or ligand coordination in binding of lead in various tissues of creosote.


Larrea tridentata (creosote bush) naturally grows and uptakes copper and lead from a contaminated area near a smelting operation. X-ray absorption spectroscopic studies conducted at Stanford Synchrotron Radiation Laboratory (SSRL) suggest copper binding to oxygen-coordinated ligands and also imply that the binding is not solely due to phytochelatins. Cu EXAFS of roots and leaves also vary depending on the level of heavy metal contamination in the environment from which the various creosote samples were obtained. In contrast, Pb XANES data of roots and leaves of creosote collected from different contaminated sites indicate no difference in valence states or ligand coordination. Based on the data obtained thus far, the binding of copper and lead ions to creosote are clearly different than that suggested by the current literature. Future research, including isolation and analyses of the various proteins, will help to elucidate the unique heavy metal binding abilities of creosote.


The research was supported through a grant by the National Institutes of Health (NIH Grant #GM 08012-25). Travel funds to SSRL were supported through a grant from GE. The authors also thank the Department of Energy, Office of Basic Energy Sciences, which funds SSRL.


Bache, C. A., W.H. Gutenmann, M. Rutzke, G. Chu, D.C. Elfving, and D.J. Lisk, 1991. Concentrations of Metals in Grasses in the Vicinity of a Municipal Refuse Incinerator, Arch. Environ. Contam. Toxicol.,20, pp.538-542.

Crist, R.H., K. Oberholser, N. Shank, and M.Nguyen, 1981. Nature of Bonding Between Metallic Ions and Algal Cell Walls, Environ. Science & Technol., 15, (10), pp.1212-1217.

Darwall, D. W., and J. Gardea-Torresdey, 1989. Removal and Selective Recovery of Heavy Metal Ions from Industrial Wastewaters, Technical Report, New Mexico Water Resources Research Institute, p.8.

Fala-Ardakan, A., 1984. Contamination of the Environment with Heavy Metals Emitted from Automotives, Exotoxicol. Environ. Safety, 8, pp.152-161.

Fernandes, J.C., and F.S. Henriques, 1991. Biochemical, Physiological, and Structural Effects of Excess Copper in Plants, The Bot. Rev., 57, pp.246-273.

Gardea-Torresdey, J. Bibb, K.J. Tiemann, J.H., Gonzalez, and J.L. Arenas, 1996a. Adsorption of Copper Ions from Solution by Heavy Metal Stressed Larrea tridentata (Creosote Bush) Biomass, L.E. Erickson, D.L. Tillison, S.C. Grant, J.P. McDonald, Eds., HSRC / WERC Conference, Proc. 11th Ann. Conf., Albuquerque, NM, pp.638-648.

Gardea-Torresdey, J., L. Polette, S. Arteaga, K.J. Tiemann, J. Bibb, and J.H. Gonzalez, 1996b. Determination of the Content of Hazardous Heavy Metals in Larrea tridentata (Creosote Bush) Grown Around a Contaminated Area, L.E. Erickson, D.L. Tillison, S.C. Grant, J.P. McDonald, Eds., HSRC / WERC Conference, Proc. 11th Ann. Conf., Albuquerque, NM, pp.660-669.

Gardea-Torresdey, J. L., M.K. Becker-Hapak, J.M. Hosea, and D.W. Darwall, 1990. Effect of Chemical Modification of Algal Carboxyl Groups on Metal Ion Binding, ES&T, 24 (9), pp.1372-1378.

Hamer, D.H., D.J. Thiele, and J.E. Lemontt, 1985. Function and Autoregulation of Yeast Copperthionein, Science, 228, pp.685-690.

Kansanen, P.H., and J. Venetvaara, 1991. Comparison of Biological Collectors of Airborne Heavy Metals Near Ferrochrome and Steel Works, Water, Air, and Soil Poll., 60, pp.337-359.

Kastori, R., M. Petrovic, and N. Petrovic, 1992. Effect of Excess Lead, Cadmium, Copper, and Zinc on Water Relations in Sunflowers, J. Plant Nutr. 15, (11) pp.2427-2439.

Lytle, C. M., F. W. Lytle, and B. N. Smith, 1996. Use of XAS to Determine the Chemical Speciation of Bioaccumulated Manganese in Potamogeton pectinatus, J. Environ. Qual., 25 (2), pp. 311-316.

Manceau, A., 1996. Direct Determination of Lead Speciation in Contaminated Soils by EXAFS Spectroscopy, E.S.&T., 30, pp.1540-1552.

Motto, H.L., R.H. Daines, D.M. Chilko, and C.K. Motto, 1970. Lead in Soils and Plants: Its Relationship to Traffic Volume and Proximity to Highways, Environ. Science & Technol. 4, (3) pp.231-237.

Narwal, R.P., M. Singh, and J.P. Singh, 1991. Effect of Nickel-Enriched Sewage Water on the Accumulation of Nickel and Other Heavy Metals in Corn, J. Indian Soc. Soil Sci., 39, pp.123-128.

Nriagu, J.O., and J.M. Pacyna, 1988. Quantitative Assessment of Worldwide Contamination of Air, Water, and Soils by Trace Metals, Nature, 333, (12) pp.134-139.

Peterson, P.J., and C.A. Girling, 1981. Other Trace Metals, In: N.W. (Ed.), Effects of Heavy Metal Pollution on Plants, vol.1, Applied Science Publishers, London, pp.213-279.

Pickering, I.J., G.N. George, C.T. Dameron, B. Kurtz, D.R. Winge, and I.G. Dance, 1993. X-ray Absorption Spectroscopy of Cuprous-Thiolate Clusters in Proteins and Model Systems, J. Amer.Chem. Soc., 115, pp.9498-9505.

Rauser, W. E. 1990. Phytochelatins, Annu. Rev. Biochem., 59, pp.61-86.

R.D. Worthington, 1989. In: Gangster, P., and H. Walter, (Eds.), Environmental Hazards and Bioresource Management in the United States-Mexico Borderlands, UCLA Latin American Center Publications, Los Angeles, pp.105-115.

Sastry, P.S., and B.R. Chaudhary, 1989. Fixation of CO2 and Incorporation of Thynidine Under Heavy Metal Stress in Closterium moniliferum, Folia Microbiol., 34, pp.106-111.

Slivinskaya, R.B. 1991. Nickel Effect on Sunflower Leaf Cell Membranes, Acta Bot. Neerl., 40, (2), pp.133-138.

Thomas, W., A. Rohling, and H. Simon, 1984. Accumulation of Airborne Pollutants (PAH, Chlorinated Hydrocarbons, Heavy Metals) in Various Plant Species and Humu, Environ. Pollut., Series A., 36, pp. 295-310.

United States EPA (April 1991) Sample Preparation for Spectrochemical Determination of Total Recoverable Elements in Biological Tissues, Method 200.3, revision 1.

United States EPA (Dec. 1987) Test Methods for the Evaluation of Solid Waste, Method 3050, Acid Digestion of Sediments, Sludges, and Soils, revision 1.

Vojtechova, M., and S. Leblova, 1991. Uptake of Cadmium and Lead by Maize Seedlings and the Effect of Heavy Metals on the Activity of Phosphoenolpyruvate Carboxylase Isolated from Maize, Bio. Plantarum (PRAHA), 33, (5) pp.386-394.

Winge, D.R., C.T. Dameron, G.N. George, I.J. Pickering, and I.G.Dance, 1993. Cuprous Thiolate Polymetallic Clusters in Biology, In: Bioinorganic Chemistry of Copper; Karlin, K.D. and Z. Tyeklar, (Eds.), Chapman and Hall, New York, pp. 110-123.

Figure 1: Copper K-edge spectra of L. tridentata roots and leaves from two different sites.

Figure 2: Copper K-edge EXAFS Fourier transforms (phase corrected for sulfur backscattering) for L. tridentata leaves compared with data for copper phytochelatin. Transform k ranges from 1 - 12A -1.

Figure 3: Copper K-edge EXAFS Fourier transforms for L. tridentata leaves from site 2and site 4. Transform k ranges from 1 - 12A -1.

Figure 4: Copper K-edge spectra of L. tridentata leaves. Site 8 leaves were reacted with 6.3 ppm Cu. Site 4 leaves were grown naturally in contaminated areas.

Figure 5:Copper K-edge EXAFS Fourier transforms for L. tridentata leaves and roots from site 4, and leaves from site 8 which were reacted with 6.3 ppm Cu.

Figure 6: Lead L III-edge spectra of leaves and roots from creosote. Also, site 8 leaves reacted with Pb are shown.