- Email: [email protected]
Phytotoxicity of coloured substances: is Lemna Duckweed an alternative to the algal growth inhibition test?
Chemosphere 49 (2002) 9–15 www.elsevier.com/locate/chemosphere
Phytotoxicity of coloured substances: is Lemna Duckweed an alternative to the algal growth inhibition test? Michael Cleuvers a
a,*
, Hans-Toni Ratte
b
Department of General Biology, Aachen University of Technology, Kopernikusstraße 16, D-52056 Aachen, Germany b Aachen University of Technology, Worringerweg 1, D-52056 Aachen, Germany Received 8 November 2001; received in revised form 23 April 2002; accepted 8 May 2002
Abstract Coloured substances cause problems when interpreting algal tests, because effects due to light absorption can interact with potential toxicity. The Lemna Duckweed growth inhibition test can complement the algal test, on condition that the test is performed on a black, not reflecting surface. On white surfaces, test solution colour can strongly impact Lemna growth. For example, average control sample growth rate of is much higher on white surfaces (0.362 d1 ) than on black surfaces (0.284 d1 ). We found that 10 mg l1 of the dyestuff ‘‘Brilliant Blue R spezial’’ inhibited average Lemna growth rate about 22% on white surfaces but did not inhibit growth on black surfaces. The reason for this difference stems from the difference in amount of light reflected from below the test beakers. With Brilliant Blue on white surfaces, the test solution colour reduces utilizable light and causes a deterioration of light conditions, whereas on a black surfaces, reflected light is absent a priori, and thus no inhibiting effect was measured. Of particular importance is the choice of test parameter. With Brilliant Blue, a LOEC for average growth rate, based on frond numbers, of 320 mg l1 was determined. However, when average growth rate was calculated using dry weights of the plants, the LOEC decreased clearly to 1.0 mg l1 . In this study, the Lemna test was much more sensitive than the algal test. We recommend Lemna tests be used in addition to algal tests, because doing so may significantly improve the assessment of phytotoxicity of chemicals and sewage. Ó 2002 Elsevier Science Ltd. All rights reserved. Keywords: Toxicity; Endpoint selection; Sensitivity; Test species; Dye
1. Introduction Due to their light absorption, coloured substances such as dyes can inhibit algal growth, which is not the result of a toxic action. This shading effect can confound measures of chemical toxicity (Cleuvers, 2001; Cleuvers et al., 2002) and thus impair the risk assessment. The European Commission (1967, 1992) recommends that such shading effects should not be taken in account. In
* Corresponding author. Tel.: +49-241-8026554; fax: +49241-8022602. E-mail address: [email protected] (M. Cleuvers).
addition to other possibilities to solve this problem, e.g. eliminating the shading effect by changing light intensity and culture volume in the algal growth inhibition test (Comber et al., 1995; Cleuvers and Ratte, 2002), the use of other primary producers as a surrogate for algal toxicity might be considered. For a few years biotests with Lemna, a free floating aquatic macrophyte (duckweed), were used in some cases to supplement or replace the algal growth inhibition test (OECD, 2000a). Green algae tolerate only a relative narrow pH-range, whereas Lemnaceans are able to grow in a wide range from pH 3.5 to 10.5. This allows testing of samples such as sewage waters, which often show unfavourable pH-values, without previous adjustment of the pH. This is of importance, because such
0045-6535/02/$ - see front matter Ó 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 2 ) 0 0 1 9 3 - 5
10
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15
adjustments can change the chemical properties and thus influence sample toxicity (Taraldsen and Norberg-King, 1990; Fomin et al., 1997). For coloured substances Lemna ssp. seemed to be a good choice, because as they float on the surface, the test solution colour should not be important. However, the possibility exists that light reflected into the beakers may play a significant role for Lemna growth. Hence, we performed tests with vessels on white and black surfaces to verify this point. Furthermore we examined several possible endpoints for their sensitivity to three tested dyes and compared this to results obtained in the algal growth inhibition test.
2. Materials and methods The tests were performed with the duckweed Lemna minor. In literature various culture mediums for Lemnaceans were described (e.g. Hutner, 1953; Landolt, 1957; Docauer, 1983). In this study we used the Steinberg-medium (Steinberg, 1943, 1946) with the modification that two kinds of phosphate were added (see Table 1) according to the standard ISO/WD 20079 (ISO, 2001). Biotests were carried out in 400 ml beakers with an outer diameter of 80 mm filled with 200 ml medium. This ensured a sufficiently large surface to allow exponential plant growth. The inoculum for each beaker was 12 fronds. Only plants with two or three fronds were chosen. As in the algal test, six control replicates and three treatment replicates were used. The test duration was seven days (168 h). Number and area of the fronds were determined at days 0, 3, 5 and 7; dry weights were measured at the end of the test. The pre-culture took place in two 5 l aquaria with the same light and temperature conditions used in the test. Tests were carried out in a climatic exposure test cabinet, calibrated at 25 2 °C, with fluorescent tubes mounted on the top. Light intensity was adjusted to 100 lEs1 m2 . To avoid differences in growth due to position in the test chamber, the site of the beakers was Table 1 Composition of the modified Steinberg-medium Substance
Concentration (mg l1 ]
KNO3 KH2 PO4 K2 HPO4 MgSO4 7H2 O CaðNO3 Þ2 4H2 O MnCl2 4H2 O H3 BO3 Na2 MoO4 2H2 O ZnSO4 7H2 O FeCl3 6H2 O Na2 EDTA 2H2 O
350 90 12.6 100 295 0.18 0.12 0.044 0.18 0.76 1.5
randomised after each measurement. The surface of the test chamber was covered with black or white paper according to the test scheme. Measurements and evaluations were performed by means of the digital image analysis system Scanalyzer produced by LemnaTec, Wuerselen, Germany. The Scanalyzer works with a height adjustable colour video camera, connected to a computer using a frame grabber (ELTEC). The image analysis software for Duckweed was produced by LemnaTec. At days 0, 3, 5 and 7 images of the beakers were taken for analysis. Following our procedures the beakers were illuminated from above, below or both. The Scanalyzer allows frond number, total frond area, mean frond size and frond size distribution to be automatically determined. Additionally, dry weight was measured. For this purpose, plants were harvested at the end of the test, dried overnight at 60 °C and weighed. Based on frond number, total frond area and dry weight, the specific average growth rates were calculated following the ISO Standard (ISO, 2001). In addition, the occurrence of chlorosis and necrosis was noted. Three dyes (Brilliant Blue R spezial, Brilliant Red 5B150 and Yellow GR), supplied by DyStar Textilfarben GmbH, Frankfurt a.M., Germany, were used as test substances. To answer the question about the impact of the surface colour, only Brilliant Blue R spezial was applied in concentrations of 10, 100 and 1000 mg l1 . For subsequent toxicity tests all dyes were tested at seven concentrations (1, 3.2, 10, 32, 100, 320 and 1000 mg l1 ). Toxicity data (NOEC’s, LOEC’s and ECx -values) were calculated with the software package EASY ASSAY using the procedures Critical Values and Multiple Testing. Variance homogeneity was tested with the Bartlett-test; the Williams-test was used for NOEC/ LOEC determination and PROBIT-analysis for the ECx -values.
3. Results Surface colour had a strong influence on growth, particularly for the controls. On white surfaces, average growth rate of frond number was 0:362 0:013 (Table 2). On black surfaces 0:284 0:031 was measured, a reduction of about 21.5%. In the various treatments difference in growth was less unequivocal. Due to stronger growth in the control, measured growth inhibitions in the white-surface treatments were much higher than on black surfaces; e.g. 10 mg l1 did not inhibit growth at all on black surfaces, whereas on white surfaces growth was reduced about 22.3% (Table 2). The impact of surface colour was even more pronounced on dry weight (Table 3). Average growth rate calculated from Lemna dry weight grown on white surfaces was
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15
11
Table 2 Average growth rates calculated from frond numbers and resulting inhibitions in the treatments as obtained in the tests with Brilliant Blue R spezial on white or black surfaces Concentration (mg l1 ) Control 10 100 1000
White surfaces
Black surfaces
Growth rate (d1 )
Inhibition (%)
Growth rate (d1 )
Inhibition (%)
0:362 0:013 0:281 0:002 0:276 0:002 0:157 0:000
– 22:29 0:65 23:69 0:67 56:65 0:00
0:284 0:031 0:291 0:000 0:230 0:029 0:178 0:019
– 2.42 0.00 19.16 10.06 37.24 6.79
Table 3 Average growth rates calculated from dry weights and resulting inhibitions in the treatments as obtained in the tests with Brilliant Blue R spezial on white or black surfaces Concentration (mg l1 )
White surfaces
Black surfaces 1
Control 10 100 1000
Growth rate (d )
Inhibition (%)
Growth rate (d1 )
Inhibition (%)
0:363 0:005 0:288 0:013 0:222 0:012 0:094 0:011
– 20:49 3:60 38:80 3:30 74:10 2:98
0:279 0:024 0:254 0:002 0:166 0:026 0:137 0:018
–
about 30% higher than on black surfaces. Of course, the black surfaces resulted in less inhibitions caused by the dye. Because the surface colour interacts with toxicity, subsequent tests with the dyes Brilliant Blue R spezial, Brilliant Red 5B150 and Yellow GR were performed on black surfaces only. There were clear differences in toxicity among the three tested dyes (Fig. 1). Yellow GR did not influence the total frond area of the plants at all; Brilliant Red inhibited frond growth rate about 18% at concentrations higher than 100 mg l1 and Brilliant Blue inhibited growth up to 33% at the highest tested concentration and was the most toxic substance, albeit the inhibitions
Fig. 1. Inhibition of the average growth rates calculated from total frond area depending on the concentration of the dyestuffs. Means and standard deviations.
8.96 0.72 40.68 9.34 50.83 6.52
Table 4 Mean frond size (% of control) of Lemna minor in dependence of the concentration of the tested dyestuffs Concentration (mg l1 )
Mean frond size (% of control) Brilliant Blue
Brilliant Red
Yellow GR
1 3.2 10 32 100 320 1000
99.4 95.8 94.0 74.9 70.7 64.7 64.7
100 86.2 108.3 103.9 78.5 81.8 82.9
104.8 104.8 106.6 100.6 109.0 103.0 102.4
Significant effect; Student’s t-test, a ¼ 0:05, two-sided.
are not very high. These results indicate that not only frond number, but also size was affected (Table 4). Particularly, Brilliant Blue strongly decreased mean frond size at concentrations higher than 3.2 mg l1 , whereas Brilliant Red reduced frond size only above 32 mg l1 . Yellow GR did not effect frond size. To study this phenomenon more in detail, frond distribution in different size classes was surveyed with Brilliant Blue R spezial (10, 100, 1000 mg l1 ; Fig. 2). Large fronds with an area >20 mm2 occurred only in controls and a few in the 10 mg l1 treatments. The controls showed the highest total frond area (2504 mm2 ) with a peak in the size class 16–18 mm2 . The higher the dye concentration, the lower the proportion of big fronds and total frond area. The peaks of the curves shifted to lower size classes. At 10 mg l1 (total area 2151 mm2 ), fronds of 10–12 mm2 were dominant, while at 100 (total area 1039 mm2 ) and 1000 mg l1 (total area 593 mm2 ) fronds of 4–6 or 6– 8 mm2 predominated.
12
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15 Table 5 Comparison of the toxicity data (NOEC, LOEC, EC10 , EC20 and EC50 ) of the tested dyestuffs calculated from different endpoints as obtained in the test with Lemna minor Average growth rate derived from:
Brilliant Blue
Brilliant Red
Yellow GR
100 320 437.1 4369.8 n.d.
3.2 10 n.d. n.d. n.d.
>1000 n.d. n.d. n.d. n.d.
Total frond area (mm2 ) NOEC 3.2 LOEC 10 EC10 25.5 EC20 177.5 7270.3 EC50
1.0 3.2 64.97 2388.3 n.d.
100 320 n.d. n.d. n.d.
Dry weight NOEC LOEC EC10 EC20 EC50
1.0 3.2 36.3 990.6 n.d.
320 1000 n.d. n.d. n.d.
Average growth rate (d1 ) (green algae) NOEC >100 >100
>100
Frond number NOEC LOEC EC10 EC20 EC50
Fig. 2. Distribution of Lemna fronds in different size classes as dependent on the concentration of Brilliant Blue R spezial.
The choice of the parameter in Lemna tests is critical. Depending on the basis of computation, different inhibitions of average growth rate caused by the dye Brilliant Blue R spezial were measured (Fig. 3). Using dry weight and total frond area yields clearly higher inhibitions than does frond number. This leads of course to differences in the effect concentrations. A comparison of these data (NOEC, LOEC, EC10 , EC20 and EC50 ) obtained with different parameters is given in Table 5. Independent of the chosen endpoint, Yellow GR was non-toxic or much less toxic than the other tested dyes. Results for Brilliant Blue R spezial and Brilliant Red 5B150 were strongly endpoint-dependent. For example, using Brilliant Blue leads to NOEC’s from 0 to 100 mg l1 , LOEC’s from 1 to 320 mg l1 and EC10 ’s from 23.5 to 437 mg l1 , depending on the chosen endpoint.
Effect-concentrations (mg l1 )
0 1.0 23.5 134.4 3799.7
For comparison, the NOEC obtained from the algal growth inhibition test is given in the last row.
For all the dyes used in this study growth rates derived from total frond area and dry weight were most sensitive and quite alike, whereas growth rates calculated from frond number demonstrated clearly fewer effects. For comparison, Table 5 also provides NOECs of the dyes obtained from the algal growth inhibition test with Scenedesmus subspicatus. Note that all dyes showed no effect at concentrations up to 100 mg l1 . Thus, the Lemna test was unequivocally more sensitive than the algal test.
4. Discussion
Fig. 3. Inhibition of the average growth rate of Lemna minor caused by Brilliant Blue R spezial as calculated from dry weight, total frond area and frond number. Means and standard deviations.
A primary reason to use Lemna in biotests with coloured substances is the fact that Lemnaceans, being part of the Pleuston, float and, hence are not affected by the colour of the test medium. However, this assumption had not been tested. Our results with Brilliant Blue R spezial have demonstrated that reflected light from the surface of the test chamber, which is reduced by coloured substances in the medium, has a distinct influence
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15
on the growth of Lemna minor. On white surfaces, the frond number and average growth rate, as well as total frond area and dry weight, were higher than on black surfaces. Because a dye in the test medium could function as a ‘‘black surface’’, it is clear that in a test with Lemnaceans the dye generates a physical effect: light conditions for the Lemnaceans are worsened by the dyestuff. Thus, with the test vessels situated on a white or bright surface, analogous to the algal test there is a physical effect due to the light absorption of the dye beyond the potential for any toxic effect. This could be shown with e.g. the average growth rate calculated from frond numbers. A concentration of 10 mg l1 Brilliant Blue caused no inhibition relative to control, if test vessels were situated on a black surface, i.e. the substance is not toxic at this level. But on white surface, a clear inhibition of about 22.3% was measured (see Table 2), which is solely attributed to the worsened light conditions in the treatments with dye. This effect is supported by the dry weight results, which was almost twice as high on white surfaces than on black surfaces, generating a 30% higher average growth rate on the white surfaces (Table 3). Interestingly this effect diminished with increasing dye concentration, until at 1000 mg l1 Brilliant Blue, average growth rate on white surface is even lower than on black surface. The reason for that is, that with increasing dyestuff concentration the light conditions in the beakers on a white surface become similar to those in the beakers on black surface. To avoid false negatives, it is mandatory that Lemna tests with dyes be performed on black, non-reflecting surfaces. Doing so ensures that the light conditions for the control plants and treatments are approximately identical and hence only toxicity and not light absorption of the dye is responsible for growth inhibition of Lemna minor. These results are the first explicit support for the benefit of using black surfaces, as prescribed in the ISO Standard ISO/WD 20079 and should be integrated too in the OECD guideline on difficult substances (OECD, 2000a). As a matter of course tests with Brilliant Blue R spezial, Brilliant Red 5B150 and Yellow GR, were performed on black surfaces. These test results were different when compared to results from the algal growth inhibition test. Yellow GR alone, caused no or only weak effects as in the algal test, but although Brilliant Red showed no effect in the test with Scenedesmus subspicatus up to the highest concentration (100 mg l1 ), with Lemna minor there was a significant inhibition of the average growth rate (calculated from dry weight or frond area) at 3.2 mg l1 . Brilliant Blue did not inhibit Lemna growth rate at 100 mg l1 , when growth rate is calculated from frond number. But considering the results derived from dry weight and total frond area, inhibition was measured at 1.0 mg l1 . This raises the question of the most suitable endpoint, which is a matter of controversy among experts regarding the
13
algal growth inhibition test (Nusch, 1982; Nyholm, 1985, 1990; Dorgerloh, 1997; Ratte et al., 1998). Table 5 compares toxicity data obtained with the different measured parameters. Particularly with Brilliant Blue, large differences are conspicuous, where total frond area and dry weight an EC10 of 25.5 and 23.5 mg l1 , respectively, were calculated, whereas with frond number a value of 437.1 mg l1 was determined. Thus, for assessing toxicity, the chosen endpoint is very important. For the dyes used in this study, no differences in classification would arise regarding the standard EUGuideline criterion (Commission of the European Communities, 1967, 1992), because of the EC50 s. Dyes used in this study have an EC50 above 100 mg l1 and thus would be labelled ‘‘non-toxic’’ for the environment. This seems to us to be a little bit careless, considering the low LOEC’s of Brilliant Blue R spezial and Brilliant Red 5B150. For other substances with higher toxicity, the choice of endpoint could be much more important for classification and labelling, which was shown in a study with potassium dichromate (Reuter, personal communication). There the EC50 calculated by frond number was 37.0 mg l1 , while with total frond area 6.1 mg l1 were determined and such a variation would cause different risk (R) phrases for classification and labelling (Commission of the European Communities, 1967, 1992). Our results indicate that a parameter which allows a better inference on biomass than frond number, such as dry weight or total frond area, should be mentioned for calculating average growth rate. Because the effect of a substance can be more complex than simply reducing frond number, e.g. a change in distribution in size classes, the use of more than one parameter is indispensable, if a serious statement about the fitness of a Lemna population is to be made. In this context it is deserving, that the OECD and ISO-Guidelines (OECD, 2000b; ISO, 2001) prescribes beneath the calculation of the frond number also the determination of dry weight or total frond area. Regarding sensitivity one could guess, that, despite of the results in this study, Lemna is less sensitive than Scenedesmus, because it is often used to remove toxicants and nutrients from sewage waters and it shows indeed a considerable tolerance against metals (Tripathi and Chandra, 1991). But this tolerance may result from an adaptation process. A general trend in sensitivity does not exist (Lewis, 1995). In a study with 16 herbicides Fairchild et al. (1997) found Lemna in eight cases to be more sensitive and ditto in eight cases to be less sensitive than the green algae Selenastrum capricornutum. Peterson et al. (1997) confirmed the findings, that green algae are not generally more or less sensitive than Lemna. In a study with eight pesticides, Lemna was either as sensitive as or more sensitive than the tested green algae (Grossman et al., 1992).
14
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15
5. Conclusions To avoid considerable faults assessing the toxicity of coloured substances, it is mandatory that Lemna tests must be performed on black, non-reflecting surfaces. Furthermore, supporting frond number an endpoint which allows an inference on biomass, such as total frond area or dry weight, should be included, if a serious statement about the fitness of a Lemna population is needed. In any case, Lemna growth inhibition tests are very useful and serve as an additional source of information about phytotoxicity for higher plants, which may differ greatly from that to algae.
Acknowledgements We would like to thank Matthias Eberius, Ilka Reuter and the whole crew of LemnaTec for technical support. Special thanks to Tom La Point for editing the manuscript for English.
References Cleuvers, M., 2001. Analyse und Bewertung des phytotoxischen Potentials farbiger Xenobiotika––Modifikationen des Algenhemmtestes mit Scenedesmus subspicatus und Erfahrungen mit alternativen Testspecies. Ph.D. thesis, Aachen University of Technology, Aachen, Germany. Cleuvers, M., Ratte, H.T., 2002. The importance of light intensity and culture volume in algal tests with coloured substances. Water Research 36 (9), 2173–2178. Cleuvers, M., Altenburger, R., Ratte, H.T., 2002. Combination effect of light and toxicity in algal tests. Journal of Environmental Quality 31 (2), 539–547. Comber, M.H.I., Smyth, D.V., Thompson, R.S., 1995. Assessment of the toxicity to algae of coloured substances. Bulletin of Environmental Contamination and Toxicology 55, 922– 928. Commission of the European Communities, 1967. Council Directive 67/548/EEC of 18th August 1967 on the approximation of the laws, regulations and administrative provisions relating to the classification, packaging and labelling of dangerous substances. Office Journal of the European Communities. L96/1. Commission of the European Communities, 1992. Council Directive 92/32EEC of 5th June 1992 amending for the 7th time Council Directive 67/548/EEC. Office Journal of the European Communities. L152. Docauer, D.M., 1983. A nutrient basis for the distribution of the Lemnaceae. Ph.D. thesis. University of Michigan, USA. Dorgerloh, M., 1997. Labor-algentest: bedeutung der toxikologischen endpunkte. Umweltwissenschaften und Schadstoffforschung 9, 222–224. European Commission, 1967. Council Directive 67/548/EEC of 18 August 1967 on the approximation of the laws, regula-
tions and administrative provisions relating to the classification, packaging and labelling of dangerous substances. Office for Official Publications of the European Communities, Luxembourg. European Commission, 1992. Council Directive 92/32/EEC of 5 June 1992 amending for the 7th time Council Directive 67/ 548/EEC. Office for Official Publications of the European Communities, Luxembourg. Fairchild, J.F., Ruessler, D.S., Haverland, P.S., Carlson, A.R., 1997. Comparative sensitivity of Selenastrum capricornutum and Lemna minor to sixteen herbicides. Archives of Environmental Contamination and Toxicology 32, 353–357. € kotoxikFomin, A., Moser, H., Pickl, C., Arndt, U., 1997. O ologische Untersuchung saurer Tagebaugew€asser der Bergbaufolgelandschaft Lausitz w€ahrend ihrer Sanierung. Hohenheimer Umwelttagung 29, 165–176. Grossman, K., Berghaus, R., Retzlaff, G., 1992. Heterotrophic plant cell suspension cultures for monitoring biological activity in agrochemicals research. Comparison with screens using algae, germination seeds and whole plants. Pesticide Science 35, 283–289. Hutner, S.H., 1953. Comparative physiology of heterotrophic Growth. In: Loomis, W.E. (Ed.), Growth and Differentiation in Plants. Iowa State College Press, pp. 417–446. ISO (International Standards Organisation) 2001. Standard ISO/WD 20079. Water quality––Duckweed growth inhibition; determination of the toxic effect of water constituents and waste water to Duckweed (Lemna minor) (Draft document). €kologische UntersuchLandolt, E., 1957. Physiologische und o ungen an Lemnaceen. Berichte der Schweizerischen Botanischen Gesellschaft 67, 271–410. Lewis, M.A., 1995. Use of freshwater plants for phytotoxicity testing––a review. Environmental Pollution 87, 319–336. Nusch, E.A., 1982. Evaluation of growth curves in bioassays. ISO/TC 147/SC 5/WG% N62. Nederlands Normalisatieinstituut, Delft, The Netherlands. Nyholm, N., 1985. Response variable in algal growth inhibition test––biomass or growth rate. Water Research 19 (3), 273– 279. Nyholm, N., 1990. Expression of results from growth inhibition toxicity tests with algae. Archives of Environmental Contamination and Toxicology 19, 518–522. OECD (Organization for Economic Cooperation, Development and Growth) 2000a. OECD Lemna growth inhibition test. OECD Guidelines for the testing of chemicals. Lemna Growth sp. Inhibition test. OECD (Organization for Economic Cooperation, Development and Growth) 2000b. Revised draft guidance document on aquatic toxicity testing of difficult substances and mixtures. Environment Directorate, Organisation for Economic Cooperation and Development. Peterson, H.G., Boutin, C., Freemark, K.E., Martin, P.A., 1997. Toxicity of hexazinone and diquat to green algae, diatoms, Cyanobacteria and duckweed. Aquatic Toxicology 39, 111–134. Ratte, H.T., Hammers-Wirtz, M., Cleuvers, M., 1998. Influence of the growth pattern on the EC50 of cell number, biomass integral and growth rate in the algae growth inhibition test. Umweltbundesamt Project Report No. 360030 10, Berlin, Germany.
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15 Steinberg, R.A., 1943. Use of Lemna as test organism. Chronica Botanica 7, 420. Steinberg, R.A., 1946. Mineral requirements of Lemna minor. Plant Physiology 21, 42–48. Taraldsen, J.E., Norberg-King, T.J., 1990. New method for determining effluent toxicty using Duckweed (Lemna
15
minor). Environmental Toxicology and Chemistry 9, 761– 767. Tripathi, B.D., Chandra, P., 1991. Chromium uptake by Spirodela polyrhiza (L.) in relation to metal chelators and pH. Bulletin of Environmental Contamination and Toxicology 47, 764–769.
Phytotoxicity of coloured substances: is Lemna Duckweed an alternative to the algal growth inhibition test? Michael Cleuvers a
a,*
, Hans-Toni Ratte
b
Department of General Biology, Aachen University of Technology, Kopernikusstraße 16, D-52056 Aachen, Germany b Aachen University of Technology, Worringerweg 1, D-52056 Aachen, Germany Received 8 November 2001; received in revised form 23 April 2002; accepted 8 May 2002
Abstract Coloured substances cause problems when interpreting algal tests, because effects due to light absorption can interact with potential toxicity. The Lemna Duckweed growth inhibition test can complement the algal test, on condition that the test is performed on a black, not reflecting surface. On white surfaces, test solution colour can strongly impact Lemna growth. For example, average control sample growth rate of is much higher on white surfaces (0.362 d1 ) than on black surfaces (0.284 d1 ). We found that 10 mg l1 of the dyestuff ‘‘Brilliant Blue R spezial’’ inhibited average Lemna growth rate about 22% on white surfaces but did not inhibit growth on black surfaces. The reason for this difference stems from the difference in amount of light reflected from below the test beakers. With Brilliant Blue on white surfaces, the test solution colour reduces utilizable light and causes a deterioration of light conditions, whereas on a black surfaces, reflected light is absent a priori, and thus no inhibiting effect was measured. Of particular importance is the choice of test parameter. With Brilliant Blue, a LOEC for average growth rate, based on frond numbers, of 320 mg l1 was determined. However, when average growth rate was calculated using dry weights of the plants, the LOEC decreased clearly to 1.0 mg l1 . In this study, the Lemna test was much more sensitive than the algal test. We recommend Lemna tests be used in addition to algal tests, because doing so may significantly improve the assessment of phytotoxicity of chemicals and sewage. Ó 2002 Elsevier Science Ltd. All rights reserved. Keywords: Toxicity; Endpoint selection; Sensitivity; Test species; Dye
1. Introduction Due to their light absorption, coloured substances such as dyes can inhibit algal growth, which is not the result of a toxic action. This shading effect can confound measures of chemical toxicity (Cleuvers, 2001; Cleuvers et al., 2002) and thus impair the risk assessment. The European Commission (1967, 1992) recommends that such shading effects should not be taken in account. In
* Corresponding author. Tel.: +49-241-8026554; fax: +49241-8022602. E-mail address: [email protected] (M. Cleuvers).
addition to other possibilities to solve this problem, e.g. eliminating the shading effect by changing light intensity and culture volume in the algal growth inhibition test (Comber et al., 1995; Cleuvers and Ratte, 2002), the use of other primary producers as a surrogate for algal toxicity might be considered. For a few years biotests with Lemna, a free floating aquatic macrophyte (duckweed), were used in some cases to supplement or replace the algal growth inhibition test (OECD, 2000a). Green algae tolerate only a relative narrow pH-range, whereas Lemnaceans are able to grow in a wide range from pH 3.5 to 10.5. This allows testing of samples such as sewage waters, which often show unfavourable pH-values, without previous adjustment of the pH. This is of importance, because such
0045-6535/02/$ - see front matter Ó 2002 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 2 ) 0 0 1 9 3 - 5
10
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15
adjustments can change the chemical properties and thus influence sample toxicity (Taraldsen and Norberg-King, 1990; Fomin et al., 1997). For coloured substances Lemna ssp. seemed to be a good choice, because as they float on the surface, the test solution colour should not be important. However, the possibility exists that light reflected into the beakers may play a significant role for Lemna growth. Hence, we performed tests with vessels on white and black surfaces to verify this point. Furthermore we examined several possible endpoints for their sensitivity to three tested dyes and compared this to results obtained in the algal growth inhibition test.
2. Materials and methods The tests were performed with the duckweed Lemna minor. In literature various culture mediums for Lemnaceans were described (e.g. Hutner, 1953; Landolt, 1957; Docauer, 1983). In this study we used the Steinberg-medium (Steinberg, 1943, 1946) with the modification that two kinds of phosphate were added (see Table 1) according to the standard ISO/WD 20079 (ISO, 2001). Biotests were carried out in 400 ml beakers with an outer diameter of 80 mm filled with 200 ml medium. This ensured a sufficiently large surface to allow exponential plant growth. The inoculum for each beaker was 12 fronds. Only plants with two or three fronds were chosen. As in the algal test, six control replicates and three treatment replicates were used. The test duration was seven days (168 h). Number and area of the fronds were determined at days 0, 3, 5 and 7; dry weights were measured at the end of the test. The pre-culture took place in two 5 l aquaria with the same light and temperature conditions used in the test. Tests were carried out in a climatic exposure test cabinet, calibrated at 25 2 °C, with fluorescent tubes mounted on the top. Light intensity was adjusted to 100 lEs1 m2 . To avoid differences in growth due to position in the test chamber, the site of the beakers was Table 1 Composition of the modified Steinberg-medium Substance
Concentration (mg l1 ]
KNO3 KH2 PO4 K2 HPO4 MgSO4 7H2 O CaðNO3 Þ2 4H2 O MnCl2 4H2 O H3 BO3 Na2 MoO4 2H2 O ZnSO4 7H2 O FeCl3 6H2 O Na2 EDTA 2H2 O
350 90 12.6 100 295 0.18 0.12 0.044 0.18 0.76 1.5
randomised after each measurement. The surface of the test chamber was covered with black or white paper according to the test scheme. Measurements and evaluations were performed by means of the digital image analysis system Scanalyzer produced by LemnaTec, Wuerselen, Germany. The Scanalyzer works with a height adjustable colour video camera, connected to a computer using a frame grabber (ELTEC). The image analysis software for Duckweed was produced by LemnaTec. At days 0, 3, 5 and 7 images of the beakers were taken for analysis. Following our procedures the beakers were illuminated from above, below or both. The Scanalyzer allows frond number, total frond area, mean frond size and frond size distribution to be automatically determined. Additionally, dry weight was measured. For this purpose, plants were harvested at the end of the test, dried overnight at 60 °C and weighed. Based on frond number, total frond area and dry weight, the specific average growth rates were calculated following the ISO Standard (ISO, 2001). In addition, the occurrence of chlorosis and necrosis was noted. Three dyes (Brilliant Blue R spezial, Brilliant Red 5B150 and Yellow GR), supplied by DyStar Textilfarben GmbH, Frankfurt a.M., Germany, were used as test substances. To answer the question about the impact of the surface colour, only Brilliant Blue R spezial was applied in concentrations of 10, 100 and 1000 mg l1 . For subsequent toxicity tests all dyes were tested at seven concentrations (1, 3.2, 10, 32, 100, 320 and 1000 mg l1 ). Toxicity data (NOEC’s, LOEC’s and ECx -values) were calculated with the software package EASY ASSAY using the procedures Critical Values and Multiple Testing. Variance homogeneity was tested with the Bartlett-test; the Williams-test was used for NOEC/ LOEC determination and PROBIT-analysis for the ECx -values.
3. Results Surface colour had a strong influence on growth, particularly for the controls. On white surfaces, average growth rate of frond number was 0:362 0:013 (Table 2). On black surfaces 0:284 0:031 was measured, a reduction of about 21.5%. In the various treatments difference in growth was less unequivocal. Due to stronger growth in the control, measured growth inhibitions in the white-surface treatments were much higher than on black surfaces; e.g. 10 mg l1 did not inhibit growth at all on black surfaces, whereas on white surfaces growth was reduced about 22.3% (Table 2). The impact of surface colour was even more pronounced on dry weight (Table 3). Average growth rate calculated from Lemna dry weight grown on white surfaces was
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15
11
Table 2 Average growth rates calculated from frond numbers and resulting inhibitions in the treatments as obtained in the tests with Brilliant Blue R spezial on white or black surfaces Concentration (mg l1 ) Control 10 100 1000
White surfaces
Black surfaces
Growth rate (d1 )
Inhibition (%)
Growth rate (d1 )
Inhibition (%)
0:362 0:013 0:281 0:002 0:276 0:002 0:157 0:000
– 22:29 0:65 23:69 0:67 56:65 0:00
0:284 0:031 0:291 0:000 0:230 0:029 0:178 0:019
– 2.42 0.00 19.16 10.06 37.24 6.79
Table 3 Average growth rates calculated from dry weights and resulting inhibitions in the treatments as obtained in the tests with Brilliant Blue R spezial on white or black surfaces Concentration (mg l1 )
White surfaces
Black surfaces 1
Control 10 100 1000
Growth rate (d )
Inhibition (%)
Growth rate (d1 )
Inhibition (%)
0:363 0:005 0:288 0:013 0:222 0:012 0:094 0:011
– 20:49 3:60 38:80 3:30 74:10 2:98
0:279 0:024 0:254 0:002 0:166 0:026 0:137 0:018
–
about 30% higher than on black surfaces. Of course, the black surfaces resulted in less inhibitions caused by the dye. Because the surface colour interacts with toxicity, subsequent tests with the dyes Brilliant Blue R spezial, Brilliant Red 5B150 and Yellow GR were performed on black surfaces only. There were clear differences in toxicity among the three tested dyes (Fig. 1). Yellow GR did not influence the total frond area of the plants at all; Brilliant Red inhibited frond growth rate about 18% at concentrations higher than 100 mg l1 and Brilliant Blue inhibited growth up to 33% at the highest tested concentration and was the most toxic substance, albeit the inhibitions
Fig. 1. Inhibition of the average growth rates calculated from total frond area depending on the concentration of the dyestuffs. Means and standard deviations.
8.96 0.72 40.68 9.34 50.83 6.52
Table 4 Mean frond size (% of control) of Lemna minor in dependence of the concentration of the tested dyestuffs Concentration (mg l1 )
Mean frond size (% of control) Brilliant Blue
Brilliant Red
Yellow GR
1 3.2 10 32 100 320 1000
99.4 95.8 94.0 74.9 70.7 64.7 64.7
100 86.2 108.3 103.9 78.5 81.8 82.9
104.8 104.8 106.6 100.6 109.0 103.0 102.4
Significant effect; Student’s t-test, a ¼ 0:05, two-sided.
are not very high. These results indicate that not only frond number, but also size was affected (Table 4). Particularly, Brilliant Blue strongly decreased mean frond size at concentrations higher than 3.2 mg l1 , whereas Brilliant Red reduced frond size only above 32 mg l1 . Yellow GR did not effect frond size. To study this phenomenon more in detail, frond distribution in different size classes was surveyed with Brilliant Blue R spezial (10, 100, 1000 mg l1 ; Fig. 2). Large fronds with an area >20 mm2 occurred only in controls and a few in the 10 mg l1 treatments. The controls showed the highest total frond area (2504 mm2 ) with a peak in the size class 16–18 mm2 . The higher the dye concentration, the lower the proportion of big fronds and total frond area. The peaks of the curves shifted to lower size classes. At 10 mg l1 (total area 2151 mm2 ), fronds of 10–12 mm2 were dominant, while at 100 (total area 1039 mm2 ) and 1000 mg l1 (total area 593 mm2 ) fronds of 4–6 or 6– 8 mm2 predominated.
12
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15 Table 5 Comparison of the toxicity data (NOEC, LOEC, EC10 , EC20 and EC50 ) of the tested dyestuffs calculated from different endpoints as obtained in the test with Lemna minor Average growth rate derived from:
Brilliant Blue
Brilliant Red
Yellow GR
100 320 437.1 4369.8 n.d.
3.2 10 n.d. n.d. n.d.
>1000 n.d. n.d. n.d. n.d.
Total frond area (mm2 ) NOEC 3.2 LOEC 10 EC10 25.5 EC20 177.5 7270.3 EC50
1.0 3.2 64.97 2388.3 n.d.
100 320 n.d. n.d. n.d.
Dry weight NOEC LOEC EC10 EC20 EC50
1.0 3.2 36.3 990.6 n.d.
320 1000 n.d. n.d. n.d.
Average growth rate (d1 ) (green algae) NOEC >100 >100
>100
Frond number NOEC LOEC EC10 EC20 EC50
Fig. 2. Distribution of Lemna fronds in different size classes as dependent on the concentration of Brilliant Blue R spezial.
The choice of the parameter in Lemna tests is critical. Depending on the basis of computation, different inhibitions of average growth rate caused by the dye Brilliant Blue R spezial were measured (Fig. 3). Using dry weight and total frond area yields clearly higher inhibitions than does frond number. This leads of course to differences in the effect concentrations. A comparison of these data (NOEC, LOEC, EC10 , EC20 and EC50 ) obtained with different parameters is given in Table 5. Independent of the chosen endpoint, Yellow GR was non-toxic or much less toxic than the other tested dyes. Results for Brilliant Blue R spezial and Brilliant Red 5B150 were strongly endpoint-dependent. For example, using Brilliant Blue leads to NOEC’s from 0 to 100 mg l1 , LOEC’s from 1 to 320 mg l1 and EC10 ’s from 23.5 to 437 mg l1 , depending on the chosen endpoint.
Effect-concentrations (mg l1 )
0 1.0 23.5 134.4 3799.7
For comparison, the NOEC obtained from the algal growth inhibition test is given in the last row.
For all the dyes used in this study growth rates derived from total frond area and dry weight were most sensitive and quite alike, whereas growth rates calculated from frond number demonstrated clearly fewer effects. For comparison, Table 5 also provides NOECs of the dyes obtained from the algal growth inhibition test with Scenedesmus subspicatus. Note that all dyes showed no effect at concentrations up to 100 mg l1 . Thus, the Lemna test was unequivocally more sensitive than the algal test.
4. Discussion
Fig. 3. Inhibition of the average growth rate of Lemna minor caused by Brilliant Blue R spezial as calculated from dry weight, total frond area and frond number. Means and standard deviations.
A primary reason to use Lemna in biotests with coloured substances is the fact that Lemnaceans, being part of the Pleuston, float and, hence are not affected by the colour of the test medium. However, this assumption had not been tested. Our results with Brilliant Blue R spezial have demonstrated that reflected light from the surface of the test chamber, which is reduced by coloured substances in the medium, has a distinct influence
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15
on the growth of Lemna minor. On white surfaces, the frond number and average growth rate, as well as total frond area and dry weight, were higher than on black surfaces. Because a dye in the test medium could function as a ‘‘black surface’’, it is clear that in a test with Lemnaceans the dye generates a physical effect: light conditions for the Lemnaceans are worsened by the dyestuff. Thus, with the test vessels situated on a white or bright surface, analogous to the algal test there is a physical effect due to the light absorption of the dye beyond the potential for any toxic effect. This could be shown with e.g. the average growth rate calculated from frond numbers. A concentration of 10 mg l1 Brilliant Blue caused no inhibition relative to control, if test vessels were situated on a black surface, i.e. the substance is not toxic at this level. But on white surface, a clear inhibition of about 22.3% was measured (see Table 2), which is solely attributed to the worsened light conditions in the treatments with dye. This effect is supported by the dry weight results, which was almost twice as high on white surfaces than on black surfaces, generating a 30% higher average growth rate on the white surfaces (Table 3). Interestingly this effect diminished with increasing dye concentration, until at 1000 mg l1 Brilliant Blue, average growth rate on white surface is even lower than on black surface. The reason for that is, that with increasing dyestuff concentration the light conditions in the beakers on a white surface become similar to those in the beakers on black surface. To avoid false negatives, it is mandatory that Lemna tests with dyes be performed on black, non-reflecting surfaces. Doing so ensures that the light conditions for the control plants and treatments are approximately identical and hence only toxicity and not light absorption of the dye is responsible for growth inhibition of Lemna minor. These results are the first explicit support for the benefit of using black surfaces, as prescribed in the ISO Standard ISO/WD 20079 and should be integrated too in the OECD guideline on difficult substances (OECD, 2000a). As a matter of course tests with Brilliant Blue R spezial, Brilliant Red 5B150 and Yellow GR, were performed on black surfaces. These test results were different when compared to results from the algal growth inhibition test. Yellow GR alone, caused no or only weak effects as in the algal test, but although Brilliant Red showed no effect in the test with Scenedesmus subspicatus up to the highest concentration (100 mg l1 ), with Lemna minor there was a significant inhibition of the average growth rate (calculated from dry weight or frond area) at 3.2 mg l1 . Brilliant Blue did not inhibit Lemna growth rate at 100 mg l1 , when growth rate is calculated from frond number. But considering the results derived from dry weight and total frond area, inhibition was measured at 1.0 mg l1 . This raises the question of the most suitable endpoint, which is a matter of controversy among experts regarding the
13
algal growth inhibition test (Nusch, 1982; Nyholm, 1985, 1990; Dorgerloh, 1997; Ratte et al., 1998). Table 5 compares toxicity data obtained with the different measured parameters. Particularly with Brilliant Blue, large differences are conspicuous, where total frond area and dry weight an EC10 of 25.5 and 23.5 mg l1 , respectively, were calculated, whereas with frond number a value of 437.1 mg l1 was determined. Thus, for assessing toxicity, the chosen endpoint is very important. For the dyes used in this study, no differences in classification would arise regarding the standard EUGuideline criterion (Commission of the European Communities, 1967, 1992), because of the EC50 s. Dyes used in this study have an EC50 above 100 mg l1 and thus would be labelled ‘‘non-toxic’’ for the environment. This seems to us to be a little bit careless, considering the low LOEC’s of Brilliant Blue R spezial and Brilliant Red 5B150. For other substances with higher toxicity, the choice of endpoint could be much more important for classification and labelling, which was shown in a study with potassium dichromate (Reuter, personal communication). There the EC50 calculated by frond number was 37.0 mg l1 , while with total frond area 6.1 mg l1 were determined and such a variation would cause different risk (R) phrases for classification and labelling (Commission of the European Communities, 1967, 1992). Our results indicate that a parameter which allows a better inference on biomass than frond number, such as dry weight or total frond area, should be mentioned for calculating average growth rate. Because the effect of a substance can be more complex than simply reducing frond number, e.g. a change in distribution in size classes, the use of more than one parameter is indispensable, if a serious statement about the fitness of a Lemna population is to be made. In this context it is deserving, that the OECD and ISO-Guidelines (OECD, 2000b; ISO, 2001) prescribes beneath the calculation of the frond number also the determination of dry weight or total frond area. Regarding sensitivity one could guess, that, despite of the results in this study, Lemna is less sensitive than Scenedesmus, because it is often used to remove toxicants and nutrients from sewage waters and it shows indeed a considerable tolerance against metals (Tripathi and Chandra, 1991). But this tolerance may result from an adaptation process. A general trend in sensitivity does not exist (Lewis, 1995). In a study with 16 herbicides Fairchild et al. (1997) found Lemna in eight cases to be more sensitive and ditto in eight cases to be less sensitive than the green algae Selenastrum capricornutum. Peterson et al. (1997) confirmed the findings, that green algae are not generally more or less sensitive than Lemna. In a study with eight pesticides, Lemna was either as sensitive as or more sensitive than the tested green algae (Grossman et al., 1992).
14
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15
5. Conclusions To avoid considerable faults assessing the toxicity of coloured substances, it is mandatory that Lemna tests must be performed on black, non-reflecting surfaces. Furthermore, supporting frond number an endpoint which allows an inference on biomass, such as total frond area or dry weight, should be included, if a serious statement about the fitness of a Lemna population is needed. In any case, Lemna growth inhibition tests are very useful and serve as an additional source of information about phytotoxicity for higher plants, which may differ greatly from that to algae.
Acknowledgements We would like to thank Matthias Eberius, Ilka Reuter and the whole crew of LemnaTec for technical support. Special thanks to Tom La Point for editing the manuscript for English.
References Cleuvers, M., 2001. Analyse und Bewertung des phytotoxischen Potentials farbiger Xenobiotika––Modifikationen des Algenhemmtestes mit Scenedesmus subspicatus und Erfahrungen mit alternativen Testspecies. Ph.D. thesis, Aachen University of Technology, Aachen, Germany. Cleuvers, M., Ratte, H.T., 2002. The importance of light intensity and culture volume in algal tests with coloured substances. Water Research 36 (9), 2173–2178. Cleuvers, M., Altenburger, R., Ratte, H.T., 2002. Combination effect of light and toxicity in algal tests. Journal of Environmental Quality 31 (2), 539–547. Comber, M.H.I., Smyth, D.V., Thompson, R.S., 1995. Assessment of the toxicity to algae of coloured substances. Bulletin of Environmental Contamination and Toxicology 55, 922– 928. Commission of the European Communities, 1967. Council Directive 67/548/EEC of 18th August 1967 on the approximation of the laws, regulations and administrative provisions relating to the classification, packaging and labelling of dangerous substances. Office Journal of the European Communities. L96/1. Commission of the European Communities, 1992. Council Directive 92/32EEC of 5th June 1992 amending for the 7th time Council Directive 67/548/EEC. Office Journal of the European Communities. L152. Docauer, D.M., 1983. A nutrient basis for the distribution of the Lemnaceae. Ph.D. thesis. University of Michigan, USA. Dorgerloh, M., 1997. Labor-algentest: bedeutung der toxikologischen endpunkte. Umweltwissenschaften und Schadstoffforschung 9, 222–224. European Commission, 1967. Council Directive 67/548/EEC of 18 August 1967 on the approximation of the laws, regula-
tions and administrative provisions relating to the classification, packaging and labelling of dangerous substances. Office for Official Publications of the European Communities, Luxembourg. European Commission, 1992. Council Directive 92/32/EEC of 5 June 1992 amending for the 7th time Council Directive 67/ 548/EEC. Office for Official Publications of the European Communities, Luxembourg. Fairchild, J.F., Ruessler, D.S., Haverland, P.S., Carlson, A.R., 1997. Comparative sensitivity of Selenastrum capricornutum and Lemna minor to sixteen herbicides. Archives of Environmental Contamination and Toxicology 32, 353–357. € kotoxikFomin, A., Moser, H., Pickl, C., Arndt, U., 1997. O ologische Untersuchung saurer Tagebaugew€asser der Bergbaufolgelandschaft Lausitz w€ahrend ihrer Sanierung. Hohenheimer Umwelttagung 29, 165–176. Grossman, K., Berghaus, R., Retzlaff, G., 1992. Heterotrophic plant cell suspension cultures for monitoring biological activity in agrochemicals research. Comparison with screens using algae, germination seeds and whole plants. Pesticide Science 35, 283–289. Hutner, S.H., 1953. Comparative physiology of heterotrophic Growth. In: Loomis, W.E. (Ed.), Growth and Differentiation in Plants. Iowa State College Press, pp. 417–446. ISO (International Standards Organisation) 2001. Standard ISO/WD 20079. Water quality––Duckweed growth inhibition; determination of the toxic effect of water constituents and waste water to Duckweed (Lemna minor) (Draft document). €kologische UntersuchLandolt, E., 1957. Physiologische und o ungen an Lemnaceen. Berichte der Schweizerischen Botanischen Gesellschaft 67, 271–410. Lewis, M.A., 1995. Use of freshwater plants for phytotoxicity testing––a review. Environmental Pollution 87, 319–336. Nusch, E.A., 1982. Evaluation of growth curves in bioassays. ISO/TC 147/SC 5/WG% N62. Nederlands Normalisatieinstituut, Delft, The Netherlands. Nyholm, N., 1985. Response variable in algal growth inhibition test––biomass or growth rate. Water Research 19 (3), 273– 279. Nyholm, N., 1990. Expression of results from growth inhibition toxicity tests with algae. Archives of Environmental Contamination and Toxicology 19, 518–522. OECD (Organization for Economic Cooperation, Development and Growth) 2000a. OECD Lemna growth inhibition test. OECD Guidelines for the testing of chemicals. Lemna Growth sp. Inhibition test. OECD (Organization for Economic Cooperation, Development and Growth) 2000b. Revised draft guidance document on aquatic toxicity testing of difficult substances and mixtures. Environment Directorate, Organisation for Economic Cooperation and Development. Peterson, H.G., Boutin, C., Freemark, K.E., Martin, P.A., 1997. Toxicity of hexazinone and diquat to green algae, diatoms, Cyanobacteria and duckweed. Aquatic Toxicology 39, 111–134. Ratte, H.T., Hammers-Wirtz, M., Cleuvers, M., 1998. Influence of the growth pattern on the EC50 of cell number, biomass integral and growth rate in the algae growth inhibition test. Umweltbundesamt Project Report No. 360030 10, Berlin, Germany.
M. Cleuvers, H.-T. Ratte / Chemosphere 49 (2002) 9–15 Steinberg, R.A., 1943. Use of Lemna as test organism. Chronica Botanica 7, 420. Steinberg, R.A., 1946. Mineral requirements of Lemna minor. Plant Physiology 21, 42–48. Taraldsen, J.E., Norberg-King, T.J., 1990. New method for determining effluent toxicty using Duckweed (Lemna
15
minor). Environmental Toxicology and Chemistry 9, 761– 767. Tripathi, B.D., Chandra, P., 1991. Chromium uptake by Spirodela polyrhiza (L.) in relation to metal chelators and pH. Bulletin of Environmental Contamination and Toxicology 47, 764–769.