Soil-Based Phytotoxicity of 2,4,6-Trinitrotoluene (TNT ... - Springer Link

Arch. Environ. Contam. Toxicol. 36, 152–157 (1999)

A R C H I V E S O F

Environmental Contamination a n d Toxicology r 1999 Springer-Verlag New York Inc.

Soil-Based Phytotoxicity of 2,4,6-Trinitrotoluene (TNT) to Terrestrial Higher Plants P. Gong,1* B.-M. Wilke,2 S. Fleischmann2 1 Department of Pollution Ecology, Institute of Applied Ecology, Chinese Academy of Sciences, P.O. Box 417, Shenyang 110015, People’s Republic of China 2 Institut fu ¨ r Landschaftsentwicklung, Technische Universita¨t Berlin, Albrecht-Thaer-Weg 4, D-14195 Berlin, Federal Republic of Germany

Received: 15 May 1998/Accepted: 30 September 1998

Abstract. Seed germination and early stage seedling growth tests were conducted to determine the ecotoxicological threshold of 2,4,6-trinitrotoluene (TNT) in two soils of different properties. Soils were amended up to 1,600 mg TNT kg21 soil and four representative species of higher plants, two dicotyledons (Lepidium sativum L., common name: cress; and Brassica rapa Metzg., turnip) and two monocotyledons (Acena sativa L., oat; and Triticum aestivum L., wheat), were assessed. Cumulative seed germination and fresh shoot biomass were measured as evaluation endpoints. Phytotoxicity of TNT was observed to be affected by soil properties and varied between plant species. Cress and turnip showed higher sensitivity to TNT than did oat and wheat. The lowest observable adverse effect concentration (LOAEC) of TNT derived from this study was 50 mg kg21 soil. In contrast to high TNT concentrations, low levels of TNT, i.e., 5–25 mg kg21 soil for cress and turnip and 25–50 mg kg21 for oat and wheat, stimulated seedling growth. Oat was capable of tolerating as much as 1,600 mg TNT kg21 and demonstrated a potential ability of TNT detoxification in one of the soils tested, suggesting that this plant might be useful in the bioremediation of TNT contaminated soils.

In recent years, phytotoxicity tests with terrestrial higher plants have been frequently used as a component of ecological risk assessment for the characterization, evaluation, and remediation process monitoring of contaminated soils. Currently available protocols for seed germination and early stage seedling growth tests (ISO 1995; Kreysa and Wiesner 1995) have been adapted from previously standardized tests, which assess the potential impacts of pesticides and other new chemical products on nontarget organisms (Fletcher 1991; Windeatt et al. 1991). In previous studies we evaluated the applicability of these soil-based bioassays for ecotoxicological assessment of contami-

nated soils (Gong et al. 1998; Wilke et al. 1998). Results have shown that the early seedling growth of oat, cress, turnip, and bush bean (Phaseolus vulgaris L.) was significantly inhibited in soils contaminated with organic pollutants (PAHs and PCBs) and/or heavy metals (Gong et al. 1998). Higher plant toxicity tests correlated very well with some soil microbial parameters, e.g., positively with nitrification and negatively with the metabolic quotient (qCO2 ) (Wilke et al. 1998). As a munitions compound 2,4,6-trinitrotoluene (TNT) is well known to be toxic, carcinogenic and mutagenic to a wide range of organisms (Klausmier et al. 1973; Won et al. 1976; Palazzo and Leggett 1986; Honeycutt et al. 1996; Peterson et al. 1996; Sunahara et al. 1998). Many studies (e.g., Funk et al. 1993; Widrig et al. 1997) have reported bioremediation of TNTcontaminated soils at ammunition sites, whereas less is known of the toxicity thresholds of TNT and its metabolites in soils. To our knowledge, only one investigation conducted by Simini et al. (1995) determined the lowest observable effect concentration (LOEC) of TNT in soil. Since soil properties may alter the bioavailability and toxicity of pollutants to higher plants (Streibig et al. 1995; Gong et al. 1998) and soil microflora (Bååth 1989), it is essential to evaluate critical levels in the presence of a soil matrix. The objectives of this study were to test the phytotoxicity of TNT spiked in soil using the seed germination and early stage seedling growth test and to compare the responses of four different species of higher plants to TNT. A secondary goal was to estimate the threshold of TNT for soil and to further evaluate the influence of soil characteristics on the phytotoxicity of TNT. In addition, oat, a species tolerant of TNT, was chosen to test its potential to detoxify TNT in soil.

Materials and Methods Soils and TNT Spiking

*Present address: Biotechnology Research Institute, National Research Council of Canada, 6100 Royalmount Avenue, Montre´al, Que´bec, Canada H4P 2R2 Correspondence to: P. Gong

Two different soils were used in the study, one collected from an experimental field at the Biologische Bundesanstalt (BBA) in Berlin, and the other provided by Landwirtschaftliche Lehr- und Faorschungsanstalt Speyer (LUFA). Both soils were screened through a 2-mm sieve and kept at field moisture until use. Their properties are shown in

Phytotoxicity of 2,4,6-Trinitrotoluene

Table 1. Appropriate amounts of TNT (technical grade, purity $ 99.9%) were first amended to 40 g acid-washed quartz sand (,0.1 mm) in porcelain evaporating dishes (90 mm in diameter). To achieve homogeneity, 15-ml acetone was added to dissolve the TNT, and a glass rod was used to stir the sand while acetone was evaporated in a ventilator. After fully drying, the TNT-amended sand was added to the soil (approximately 3.5 kg dry weight) and then mixed with an electric household blender for 15 min. For the control, 40 g nonamended sand was mixed with the soil. The designed TNT gradient was 0, 25, 50, 100, 200, 400, 800, and 1,600 mg kg21 dry soil.

153

Table 1. Some selected physicochemical properties of the soils tested

Soil

pH Corg WHC (CaCl2 ) (%) (%)

BBA 6.6 LUFA 5.6

2.5 26 2.5 45

K P

Nmin NTotal Texture (%)

(mg 100 g21 soil) 37 54 0.20 168 19 0.9 0.13 206

Clay Silt Sand 5.3 7

24 16

71 77

Plant Species and Seed Viability Examination Four species of higher plants were used for testing: two monocotyledons (Acena sativa L., common name: oat; and Triticum aestivum L., wheat) and two dicotyledons (Lepidium sativum L., cress; and Brassica rapa Metzg., turnip). Seeds of these species were purchased from commercial sources. Before use their germination potential was examined at 20 6 1°C in the dark. Germination rates over 90% guaranteed the viability of the seeds.

Fig. 1. Schematic diagram of the self-watering test system modified after Stalder and Pestemer (1980)

Testing Facilities The seed germination and early seedling growth tests were carried out in a greenhouse, which was maintained at the following conditions: lighting 16 h; light intensity ca 17,000 lux; temperature 21/17 6 1°C (day/night); relative humidity 40–60%. An automatic watering system (Figure 1), which was adopted from Stalder and Pestemer (1980) and evaluated and refined in our previous studies (Gong et al. 1998), was employed to avoid the time-consuming daily manual adjustment of soil moisture. Moist soil of 200 g equivalent oven-dry weight was placed in a plastic pot (diameters: 9 cm [top] and 6.5 cm [bottom]; height: 6.5 cm) together with a thread of glass fiber (length: 8 cm; diameter: 1 mm). A petri dish and a plastic saucer were placed under the pot, and distilled water (stored overnight in the greenhouse) was refilled to continuously maintain available water in the petri dish as needed.

Chemical Analysis of TNT About 20 g freeze-dried soil was g21 extracted for 6 h with 70 ml hexane: benzene (2:1) using a Soxhlet extraction apparatus. The extract was filtered and then reduced to 1 ml in a stream of nitrogen gas. Two microliters concentrated extract was injected onto a GC (HP 5890 Series II) equipped with a MSD (HP 5971 A) and an Optima 5 column (25 m 3 0.35 µm). The injector temperature was set at 300°C, and the programming for the oven was 5 min at 80°C followed by 10°C min21 until 280°C. All measurements were performed in SIM-Mode (m/ z 5 210). The chemical analyses were carried out by AZB (Analytical Center Berlin)-Adlershof GmbH, Berlin, Germany.

Data Analysis Test Procedure and Experimental Design Ten seeds per pot were sown uniformly in the soil to a depth of about 1.5 cm (for oat and wheat) or 0.5 cm (for cress and turnip) as depicted in Figure 1. After 50% of the seeds had germinated in the control soils, the seedlings were thinned to leave five most uniform per pot, and the test was terminated 14 days later. The seedling shoots were then cut above the soil surface and the total fresh biomass of each pot (five shoots or less) was weighed. Cumulative germination was recorded daily until the termination of the test. Two experiments were performed with four replicates per treatment, and pots were randomly placed on a bench in the greenhouse. In the first experiment, lower concentrations of TNT were used and all four species were tested following the procedure described above. A lower gradient of 0, 25, 50, 100, and 200 mg TNT kg21 soil was designed for cress and turnip, whereas a relatively higher gradient of 0, 50, 100, 200, and 400 mg TNT kg21 soil for oat and wheat. In the second experiment, the BBA and the LUFA soils were both amended with 0, 800, and 1,600 mg TNT kg21 dry soil and only oat tested. Half of the pots were sown with oat seeds, while the other half were left standing in the greenhouse without sowing. After a first harvest, oat roots were removed from the soil. All pots were left for another 2 weeks in the greenhouse under the same conditions as during the test (e.g., watering and lighting). Afterward, oat seeds were sown again into all the pots. The next harvest was conducted 14 days after 50% germination in the control pots.

Both the cumulative germination rate (total number of germinated seeds as a percentage of the number of seeds sown in each pot) and the fresh biomass of five shoots per pot were taken as the measurement endpoints. In both experiments, there were pots having less than five seeds germinated (especially at high TNT doses), and there were also some cases in which more than five seeds germinated but less than five living seedlings remained in a pot at the time of harvest. Therefore, the biomass was normalized in order to compensate for the loss of total biomass, which was due to either the poor germination or the selection of unhealthy seedlings at the time of thinning. Data normalization was done using the following equation: B n 5 Ba 3 5 / N

where Bn 5 normalized fresh biomass per pot; Ba 5 actually harvested fresh biomass; and N 5 number of shoots left at harvest. Experimental results were presented and statistically analyzed after being normalized. All data were subject to a two-tailed t test or analysis of variance (ANOVA) using the ANOVA/MANOVA module of the STATISTICA (release 5) statistical package (StatSoft Inc. 1996). Post hoc comparisons (least significant difference [LSD] test) were performed to examine significant difference between means setting a 5 0.05.

154

Results and Discussion TNT Recovery

P. Gong et al.

Table 2. TNT (2,4,6-trinitrotoluene) recovered from soil samples after amendment. Values of the determined concentrations are given in averages of measurements of two subsamples for each treatment BBA

The actual concentrations of TNT measured in soil samples immediately after blending were lower than anticipated (the two highest levels were not determined). As seen in Table 2, the recovery of TNT increased with the amendment concentration, which was probably due to the loss associated with the TNT spiking. Based on the data in Table 2, it can be calculated that TNT losses did not exceed 60 mg kg21 soil (or 210 mg for 3.5 kg soil) except for the 400 mg kg21 amendment in the LUFA soil. For each amendment level losses were similar for both the BBA and the LUFA soils. Therefore, it was likely that the first step, i.e., dissolving TNT with acetone and evaporating in order to adsorb TNT to quartz sand, primarily caused the major losses. This can be explained by the fact that the porcelain evaporating dishes of the same size were used in this step. The maximum amount of TNT that adsorbed to the internal walls of these utensils should be a constant figure, likely less than 210 mg TNT in this case, considering that the same amounts of sand and acetone were used. It is thus rational that so low recovery rates (,50%) were found in the 25, 50, and 100 mg kg21 treatments. Moreover, other steps of the spiking procedure (e.g., volatilization, translocation of TNT spiked sand) may also have contributed to the TNT losses.

Fresh Shoot Biomass The fresh biomass of 14-day early seedlings harvested in the first experiment is shown in Figure 2. Cress and turnip exhibited higher sensitivity to TNT than did oat and wheat. This is consistent with the results from an early study (Gong et al. 1998). However, except for wheat at about 50 mg kg21, at the two lowest spiking concentrations (,55 mg kg21 for oat and wheat, while ,25 mg kg21 for cress and turnip, see Table 2), stimulatory effects were observed when comparing the averages of the fresh biomass. Statistically, this stimulation was significant for oat as well as wheat at 20 mg kg21 in the LUFA soil (Figure 2). Therefore, the lowest observable adverse effect concentration (LOAEC) of TNT was 50 mg kg21. The responses of cress and turnip to the same TNT concentrations differed markedly between the two tested soils. Since significant inhibition was observed at a lower TNT level (50 mg kg21 ) in the BBA soil (Figure 2), TNT was more toxic to cress and turnip in the BBA soil than in the LUFA soil. This was probably due to the lower pH and higher fertility of the LUFA soil (Table 1). Streibig et al. (1995) reported a negative correlation between ED50 (effective dose for 50% reduction) of incorporated metsulfuron-methyl and soil pH when they analyzed results from 74 independently run bioassays using cress, turnip, and sugarbeet (Beta vulgaris). A soil pH of 5.6 was probably more preferable to these two dicotyledons. Higher fertility or productivity of the LUFA soil, which was associated with higher production of three (oat, cress, and turnip) out of all four species in the LUFA soil other than the BBA soil, might also mask the toxicity of TNT. This had been reported in our previous studies (Gong et al. 1998).

LUFA

Designed Actual Actual Concentration Concentration Recovery Concentration Recovery (mg kg21 soil) (mg kg21 soil) (%) (mg kg21 soil) (%) 0 25 50 100 200 400

0 7.8 24 54 158 355

31 48 54 79 89

0 5 20.5 49 136 311

20 41 49 70 78

Seed Germination Germination was less sensitive to TNT toxicity than was seedling growth. As shown in Figure 3, up to 158 and 350 mg TNT kg21 did not affect the cumulative emergence of turnip and oat or wheat in the BBA soil, respectively. Similar effects were observed in the LUFA soil (data not shown). On the other hand, a reduction in shoot biomass was observable at much lower TNT concentrations (Figure 2). Seed germination depends on the energy reserves in cotyledons, which makes it less sensitive to environmental pollutants (Gong et al. 1998). Besides, it was difficult to identify germinated seeds during the first several days, especially for small seeds like cress and turnip, because these seeds were lying beneath soil. The seed germination alone therefore is not a good indicator of phytotoxicity. However, we believe that invisible damages occurring during seed germination could be accumulative and may become apparent during the following seedling growth stage. In fact, the measured effects on shoot biomass should be considered as the cumulative effects during both the seed germination and the early seedling growth stages. However, when present at high concentrations, TNT can not only retard seedling growth but also extinguish seed germination of sensitive species such as cress. According to Peterson et al. (1996), germination and root and shoot growth rates (length per day) of tall fescue decreased linearly with increasing TNT concentrations, and exposure to 30 mg TNT L21 or greater consistently reduced germination and inhibited seedling development. In this study, we found that cress and turnip could germinate only at , 200 mg TNT kg21 (data not shown).

Detoxification as Evidenced by the Higher Plant Bioassay The second experiment was designed to look at the detoxification of high TNT doses by oat, the most tolerant species among the four tested. It was not surprising that the seedling growth of oat was retarded substantially by 800 and 1,600 mg TNT kg21 (Table 3). Although the germination of oat seeds was still not affected by TNT (data not shown), we observed that oat seedlings grown in TNT-amended soils were significantly shorter than those in the control soils. Stunted root development such as sparser, shorter, and abnormal root hairs was also observed in the TNT treated samples. Peterson et al. (1996) reported similar morphological influence of TNT on tall fescue (Festuca arundinacea Schreb.).


155

Fig. 2. Fresh biomass of early seedling shoots of four higher plant species grown in two soils: (a) BBA and (b) LUFA. Means (points) and standard deviations (SDs, error bars) were calculated from four replicates. Asterisks indicate significant difference from controls at 95% (*) or 99% (**) confidence (t test)

Fig. 3. Cumulative emergence of plant seedlings at the terminal day of testing in the BBA soil. Means (columns) and SDs (error bars) were obtained from four replicates (each 10 seeds). Different letters over error bars indicate significant difference at p , 0.05 (ANOVA, LSD test)

Evidence of TNT detoxification was found in the BBA soil, possibly through microbial transformation and mineralization and plant uptake. Compared to the first harvest (oat-1), the second harvest of oat shoots (oat-2) was significantly im-

roved in the BBA soil for all TNT treatments, but not for the controls (Table 3), which could be attributed to detoxication of TNT in this soil. The insignificant difference between the control and the 800 mg TNT kg21 treatment for both

156

P. Gong et al.

Table 3. Oat shoot biomass harvested in soil spiked with high concentrations of TNT (800 and 1,600 mg kg21 ) in comparison with that in the control Soil/ TNT Fresh Harvest Treatmenta Biomass (g)b BBA Oat-1

Oat-2

Oat-18

LUFA Oat-1

Oat-2

Oat-18

1 2 3 1 2 3 1 2 3

0.964 6 0.112 0.610 6 0.044 0.620 6 0.109 1.065 6 0.152 0.886 6 0.144 0.765 6 0.039 1.066 6 0.141 0.784 6 0.241 0.747 6 0.102

1 2 3 1 2 3 1 2 3

1.066 6 0.164 0.484 6 0.083 0.460 6 0.074 0.771 6 0.085 0.402 6 0.076 0.293 6 0.081 0.823 6 0.070 0.395 6 0.091 0.299 6 0.105

Oat-1 1

2

Oat-2 3

* * — — * * — —

1

2

— * — — —

—

* * *

Oat-18 3

1

2

3

— — * —

— — *

* — *

* * — — —

* — *

—

Oat-1, Oat-2, and Oat-18 stand for the first and second harvest from identical pots without prestoring, and the first harvest from pots stored for 1 month before testing, respectively. Significant difference between two means is marked * whereas insignificant difference is — (p , 0.05, t test) a 1 5 control; 2 5 800 mg TNT kg21; 3 5 1,600 mg TNT kg21 b Normalized fresh weight of five seedling shoots per pot (mean 6 SD, n 5 4)

oat-2 and oat-18 (Table 3) also suggests that TNT was detoxified. It was anticipated that TNT concentrations should decline substantially after the first harvest, especially after the removal of oat roots, and during the subsequent 2-week storage in the greenhouse. It has been demonstrated that TNT can be transformed and mineralized by microorganisms indigenous to the soils at an inactive munitions plant (Bradley et al. 1994). Uptake by higher plants is also a possible pathway. In hydroponic cultures containing TNT, Palazzo and Leggett (1986) found TNT uptake by yellow nutsedge (Cyperus esculentus L.), and Hughes et al. (1997) further confirmed the intrinsic ability of aquatic plants (Myriophyllum spicatum and M. aquaticum) to uptake and transform TNT. Schneider et al. (1996) observed that TNT accumulated in the roots of plants growing in TNT-contaminated soils. In the LUFA soil, however, no evidence of TNT detoxification could be found in relation to soil storage and the initial harvest. In contrast to that in the BBA soil, the second harvest in the LUFA soil was lower than the first for both TNT-treated and control samples. This might be caused by nutrient deficiencies or depletion in soil as significant decrease in fresh biomass took place in the controls as well. Addition of a nutrient solution was recommended for soils having low fertility (Kreysa and Wiesner 1995). However, attempts of nutrient treatment failed to identify the presence of phytotoxic substances in soils, which had been confirmed by chemical analyses (Sirois, 1990; Gong

et al. 1998). Other factors such as poor physical structure and pathogenic infection may also partially explain the decrease of fresh biomass from oat-1 to oat-18 and oat-2 in this soil (Table 3). In this study, significant difference existed between the control and 1,600 mg TNT kg21 for all three harvests in both BBA and LUFA soils (Table 3), implying that microbial TNT mineralization or TNT disappearance by plant uptake may be inhibited by high TNT concentrations (Bradley and Chapelle 1995). TNT metabolites such as 4-amino-2,6-dinitrotoluene, 2-amino-4,6-dinitrotoluene, and 2,6-diamino-4-nitrotoluene have been shown to be toxic and may exert inhibitory effects on higher plants (Peterson et al. 1996), which makes it difficult to identify the mechanisms underlying the loss of TNT and its relationship to the reduction in toxicity exerted by TNT and its metabolites. However, Simini et al. (1995) reported that TNT and 1,3,5-trinitrobenzene accounted for the toxicity of soils at two ammunition sites and that biodegradation products of TNT contributed little to soil toxicity. Won et al. (1976) also found that the major microbial metabolites of TNT were nontoxic and nonmutagenic. Since TNT metabolites were not analyzed in this study, no data are available to distinguish the phytotoxicity of TNT from that of its metabolites.

Conclusion This study demonstrated that low concentrations of TNT stimulated while high concentrations inhibited the growth of all four species of higher plants in both the BBA and LUFA soils. Cress and turnip were more sensitive to TNT phytotoxicity than were oat and wheat. The inhibitory LOEC of TNT was about 50 mg kg21 for cress and turnip in the BBA soil. Soil itself is an important factor that should be taken into account when performing soil-based bioassays. Findings from this as well as previous studies confirmed the influence of soil chemical, physical, and biological properties on bioassay results (Sirois 1990; Simini et al. 1995; Gong et al. 1998). Detoxification was observed when oat was growing in the BBA soil, which suggests that tolerant plants can play an important role in the bioremediation of TNT contaminated soil. According to Anderson et al. (1993), vegetation can enhance microbial degradation rates of organic chemical residues in soils, possibly by offering a favorable rhizospheric environment to soil microbial communities. The seed germination and early seedling growth bioassay is cost-effective, rapid, and sensitive for identifying phytotoxic substances in soil. Although the seed germination alone is not a sensitive indicator, the shoot biomass of early seedlings that reflects both effects of TNT on germination and seedling growth proved to be a sensitive measurement endpoint.

Acknowledgments. This work was sponsored by BMBF (Federal Ministry for Education, Science, Research, and Technology) of Federal Republic of Germany under Contract No. FKZ1491031. P. Gong received a BMBF Minister’s scholarship. We would like to thank Dr. P. Ko¨hler with AZB (Analytical Center Berlin)-Adlershof GmbH for TNT analyses and Drs. E. Witter, G. I. Sunahara, and S. D. Siciliano for their critical comments and for improving the text.


References Anderson TA, Guthrie EA, Walton BT (1993) Bioremediation in the rhizosphere. Environ Sci Technol 27:2630–2636 Bååth E (1989) Effects of heavy metals in soil on microbial processes and populations (a review). Water Air & Soil Pollut 47:335–379 Bradley PM, Chapelle FH (1995) Factors affecting microbial 2,4,6trinitrotoluene mineralization in contaminated soil. Environ Sci Technol 29:802–806 Bradley PM, Chapelle FH, Landmeyer JE, Schumacher JG (1994) Microbial transformation of nitroaromatics in surface soils and aquifer materials. Appl Environ Microbiol 60:2170–2175 Fletcher J (1991) A brief overview of plant toxicity testing. In: Gorsuch JW, Lower WR, Wang W, Lewis MA (eds) Plants for toxicity assessment, 2nd vol. ASTM STP 1115, American Society for Testing and Materials, Philadelphia, PA, pp 5–11 Funk SB, Roberts DJ, Crawford DL, Crawford RL (1993) Initial-phase optimization for bioremediation of munition compound-contaminated soils. Appl Environ Microbiol 59:2171–2177 Gong P, Wilke B-M, Strozzi E, Fleischmann S (1998) Higher plant seed germination and early seedling growth test: evaluation and refinement for the use in ecotoxicological assessment of soils. In: Proceedings of the International Symposium on Soil, Human and Environment Interactions, May 4–11, 1997, Nanjing, China (in press) Honeycutt ME, Jarvis AS, McFarland VA (1996) Cytotoxicity and mutagenicity of 2,4,6-trinitrotoluene and its metabolites. Ecotoxicol Environ Safety 35:282–287 Hughes JB, Shanks J, Vanderfor M, Lauritzen J, Bhadra R (1997) Transformation of TNT by aquatic plants and plant tissue cultures. Environ Sci Technol 31:266–271 ISO (International Organization for Standardization) (1995) Soil quality—determination of the effects of pollutants on soil flora, Part 2: effects of chemicals on the emergence and growth of higher plants. ISO 11269–2. Geneva, Switzerland Klausmier RE, Osmon JL, Walls DR (1973) The effect of trinitrotoluene on microorganisms. Develop Indus Microbiol 15:309–317 Kreysa G, Wiesner J (1995) Bioassays for soils. DECHEMA (Deutsche Gesellschaft fu¨r Chemisches Apparatewesen, Chemiche Technik und Biotechnologie) e.V., Frankfurt, pp 22–24 Palazzo AJ, Leggett DC (1986) Effect and disposition of TNT in a terrestrial plant. J Environ Qual 15:49–52

157

Peterson MM, Horst GL, Shea PJ, Comfort SD, Peterson RKD (1996) TNT and 4-amino-2,6-dinitrotoluene influence on germination and early seedling development of tall fescue. Environ Pollut 93:57–62 Schneider K, Oltmanns J, Radenberg T, Schneider T, Pauly-Mundegar D (1996) Uptake of nitroaromatic compounds in plants: implications for risk assessment of ammunition sites. Environ Sci Pollut Res 3:135–138 Simini M, Wentsel RS, Checkai RT, Philips CT, Chester NA, Major MA, Amos JC (1995) Evaluation of soil toxicity at Joliet Army Ammunition Plant. Environ Toxicol Chem 14:623–630 Sirois DL (1990) Survey of Rocky Mountain Arsenal for phytotoxic substances. In: Wang W, Gorsuch JW, Lower WR (eds) Plants for toxicity assessment. ASTM STP 1091, American Society for Testing and Materials, Philadelphia, PA, pp 188–197 Stalder L, Pestemer W (1980) Availability to plants of herbicide residues in soil. Part I: A rapid method for estimating potentially available residues of herbicides. Weed Res 20:341–347 StatSoft Inc. (1996) STATISTICA for Windows (computer program manual). Tulsa, OK Streibig JC, Walker A, Blair AM, Anderson TG, Eagle DJ, Friedlander H, Hacker E, Iwanzik W, Kudsk P, Labhart C, Luscombe BM, Madafiglio G, Nel PC, Pestemer W, Rahman A, Retzlaff G, Rola J, Stefanovic L, Straathof HJM, Thies EP (1995) Variability of bioassays with metsulfuron-methyl in soil. Weed Res 35:215–224 Sunahara GI, Dodard S, Sarrazin M, Paquet L, Ampleman G, Thiboutot S, Hawari J, Renoux AY (1998) Development of a soil extraction procedure for ecotoxicity characterization of energetic compounds. Ecotoxicol Environ Safety 39:185–194 Widrig DL, Boopathy R, Manning JF Jr (1997) Bioremediation of TNT-contaminated soil: a laboratory study. Environ Toxicol Chem 16:1141–1148 Wilke B-M, Gong P, Papassotiriou J, Strozzi E, Fuchs M, Fleischmann S (1998) Bioassays for soils: evaluation of higher plant growth and microbial toxicity tests. In: Proceedings of the International Symposium on Soil, Human and Environment Interactions, May 4–11, 1997, Nanjing, China (in press) Windeatt AJ, Tapp JF, Stanley RD (1991) The use of soil-based plant tests based on the OECD guidelines. In: Gorsuch JW, Lower WR, Wang W, Lewis MA (eds) Plants for toxicity assessment, 2nd vol. ASTM STP 1115, American Society for Testing and Materials, Philadelphia, PA, pp 29–40 Won WD, DiSalva LH, Ng J (1976) Toxicity and mutagenicity of 2,4,6-trinitrotoluene and its microbial metabolites. Appl Environ Microbiol 31:576–580