Toxicity of Sodium Molybdate and Sodium Dichromate to Daphnia Magna Straus Evaluated in Acute, Chronic, and Acetylcholinesterase Inhibition Tests Teresa C

Ecotoxicology and Environmental Safety 45, 253}259 (2000) Environmental Research, Section B doi:10.1006/eesa.1999.1889, available online at http://www.idealibrary.com on

Toxicity of Sodium Molybdate and Sodium Dichromate to Daphnia magna Straus Evaluated in Acute, Chronic, and Acetylcholinesterase Inhibition Tests Teresa C. Diamantino,* - Lucia Guilhermino,? Elisabete Almeida,- and Amadeu M. V. M. Soares* *Instituto do Ambiente e Vida, Departamento de Zoologia, Largo Marque' s de Pombal, Universidade de Coimbra, 3000 Coimbra, Portugal; -Instituto Nacional de Engenharia e Tecnologia Industrial, Estrada do Pac,o do Lumiar, 1699 Lisboa Codex, Portugal; and ?Instituto de Cie' ncias BimeH dicas de Abel Salazar, Departamento de Estudos de Populac,oJ es, LaboratoH rio de Ecotoxicologia/Centro de Investigac,a o Marinha e Ambiental, Universidade do Porto, Lg Prof. Abel Salazar, 2, 4050 Porto, Portugal Received March 18, 1999

spread application in numerous industrial processes, chro- As a result of a widespread application in numerous industrial mium is a contaminant of many environmental systems processes, chromium is a contaminant of many environmental systems. Chromium and their compounds are toxic to both (Cohen et al., 1993). Industrial uses of chromium include invertebrates and vertebrates and, for this reason, there has been metal "nishing industry, production of paints and pigments, a search for suitable and less toxic alternatives. Molybdenum tanning, wood preservation, chromium chemicals produc- compounds have been studied as alternative to chromium tion, pulp and paper production, and metal smelting (Paw- compounds for some industrial applications. The toxicity of lisz et al., 1997). One of the major uses of environmentally chromium is well known but the e4ects of molybdenum and hazardous chemicals is in the metal "nishing industry. So- molybdenum mining on natural populations and communities of dium dichromate is the most important of the industrial freshwater invertebrates have not often been studied. However, chromium chemicals; it is used as the starting compound for chromium, and molybdenum (and their compounds) are included almost all chromium compounds (Anger et al., 1986) and is in the same list (List II) of European Union dangerous substan- "rmly established in the metal "nishing "eld (Biestek and ces. In this study, the acute and chronic e4ects of sodium molyb- Weber, 1976). One of the environmental problems of these date and sodium dichromate to Daphnia magna Straus were evaluated. Furthermore, in vitro and in vivo e4ects of these two industries is the elimination of hexavalent chromium com- metals on acetylcholinesterase (AChE) activity of D. magna pounds, which are used extensively for the passivation of

Straus were investigated. LC50 values determined at 48 h were zinc, and for zinc-based coatings and surfaces. The elimina- -and 2847.5 mg L؊1 for chromium (as sodium dichromate) tion of these toxic chemicals from metal processing, parti 0.29 and molybdenum (as sodium molybdate), respectively. No signif- cularly in the automobile industry where chromates are icant in vitro e4ects of both metals on AChE were found. How- widely used for the protection of zinc-plated parts, is con- ever, both toxicants inhibited AChE in vivo at concentrations sidered a priority within the European Union (SI, 1997). under the respective 48-h LC50 values. Both sodium dichromate Molybdenum compounds are used in catalysis, luber"z- and sodium molybdate inhibited the reproduction and growth of ation, #ame retardancy and smoke suppression, pigments, D. magna, but the concentrations inducing signi5cant e4ects agriculture, and corrosion inhibition (Sebenik et al., 1990). were di4erent for the two chemicals. Sodium molybdate had At high concentrations, molybdenum is toxic to many or- signi5cant lower toxicity to D. magna Straus than sodium dich- ganisms, including mammals and freshwater invertebrates romate. ( 2000 Academic Press Key Words: sodium molybdate; sodium dichromate; acute and (Khangarot, 1991). The e!ects of molybdenum or molyb- chronic e4ects; AChE, Daphnia magna. denum mining on natural populations or communities of freshwater invertebrates have not often been studied (Whiting et al., 1994). Based on toxicological studies from 1. INTRODUCTION the literature, some authors have been studying alternatives to chromium hexavalent compounds for passivation of zinc Heavy metals are widely recognized as highly toxic and coatings based on molybdenum compounds (Tang et al, dangerous to organisms (Mance, 1987). As a result of wide- 1994; Wharton et al., 1996; Almeida et al., 1998). Despite the di!erences of toxicity between chromium and molybdenum To whom correspondence should be addressed at IMP/LTR, Instituto reported in the literature, both metals (and their com- Nacional de Engenharia e Tecnologia Industrial, Estrada do Pac,odo Lumiar, 1699 Lisboa Codex, Portugal. Fax: #351-21-7160901. E-mail: pounds) are included in the List II of European dangerous [email protected]. substances directive (CEC, 1976).

253

0147-6513/00 $35.00 Copyright ( 2000 by Academic Press All rights of reproduction in any form reserved. 254 DIAMANTINO ET AL.

Daphnia magna is widely used as a test organism in test was carried out for 48 h. The criterion of toxic e!ect was aquatic toxicology (Adema, 1978; Meer et al., 1988; Soares death, recognized by immobilization after stimulation by et al., 1992). Bioassays involving this species are required for a bright light. In chronic tests, the organisms were cultured the assessment of the potential impact of new chemicals on in ASTM medium with an organic additive (Baird et al., the aquatic environment (Soares et al., 1991). Acute and 1989), one animal per 1000 ml of medium, and fed with the chronic bioassays with this species are the most widely used algae C. vulgaris (0.322 mg carbon/daphnia/day). Each test test methods for the toxicity assessment of chemical com- was carried out for 21 days. Animals were transferred to pounds and e%uents. In acute tests the measured parameter newly prepared test solutions three times a week (i.e., every is death, while in chronic tests the inhibition of normal other day). One control and six toxicant concentrations reproduction and growth are the most used endpoints for with 10 replicates for each treatment were used. Endpoints toxicity evaluation. However, chronic tests are particularly were reproduction, growth, and mortality. In both acute time-consuming and expensive. For this reason, in recent and chronic tests, temperature and photoperiod were as decades there have been e!orts to develop and validate described above. alternative methods to conventional toxicity tests (Blaise et al., 1997; Tylor et al., 1998). 2.2. In Vitro AChE Inhibition Test Enzymes are known to play a crucial role in vital func- tions of all organisms. The inhibition of cholinesterases has Tests were performed as described in Guilhermino et al. been widely used as a speci"c biomarker for organophos- (1996) with the following modi"cations: animals were cul- phate and carbamate pesticides, being considered a &&gold tured for 48 h in the presence of food; homogenates were standard biomarker'' (Peakall, 1994). However, studies pub- stored at !203C for a maximum of 1 week. The incubation lished in recent years indicate that some metals (including procedure was performed according to Herbert et al. chromium) might also inhibit these enzymes, under both in (1995/1996). In each experiment, homogenate samples vivo and/or in vitro conditions (Labrot et al., 1996; Payne et (495 ll) were incubated at 203C with 5 ll of each test solu- al., 1996; Guilhermino et al., 1998). tion, for 30 min. For each compound, a water control and In this study, the toxicity of sodium dichromate and "ve concentrations of the metallic element were used (12.5, sodium molybdate to D. magna was evaluated using conven- 25, 50, 75, and 100 mg L\). Three replicates per treatment tional 48-h acute tests, in vivo and in vitro acetylcholines- were used. The activity of AChE was determined, in tripli- terase (AChE) inhibition assays, and 21-day chronic tests. cate, by the Ellman method (Ellman et al., 1961) adapted to Furthermore, the e!ect of the color of sodium dichromate in microplate (Herbert et al., 1995/1996). The protein content the results of an in vitro AChE inhibition test was also of the samples was determined according to the Bradford investigated. technique (Bradford, 1976) with the modi"cations described by Herbert et al. (1995/1996). The activity of AChE in each sample was presented as the mean of the three determina- 2. MATERIALS AND METHODS tions performed and expressed in units (U) per milligram of protein (1 U"1 nmol of substrate hydrolyzed per In the following sections total chromium and molyb- minute). A Labsystems Multiskan MS microplate reader denum concentrations refer to the concentrations of the was used. metallic element in the medium. Sodium dichromate solutions presented an orange color Parent animals were cultured in ASTM hard water interfering with both AChE and protein determinations. (ASTM, 1980) with an organic additive (Baird et al., 1989), For this reason, a correction test for color interference as in groups of 10 animals per 1000 ml of medium, and fed with described above but without Daphnia homogenate was per- the algae Chlorella vulgaris (0.322 mg carbon/daphnia/day). formed in parallel. The objective of this procedure is to The photoperiod was 16 h light : 8 h dark and the temper- account for eventual e!ects due to the color of the chro- ature was 20$13C. mium solutions.

2.1. Acute and Chronic Tests 2.3. In Vivo AChE Inhibition Tests Tests were carried out in ASTM hard water (ASTM, Tests were carried out according to Guilhermino et al. 1980), with animals from a single clone (clone A, sensu; (1996) with the following modi"cations: 60 animals were Baird et al., 1989) and initiated with third- to "fth-brood used per treatment, in groups of 20 per 1000 ml of test neonates ((24 h). In acute tests, organisms were not fed solution. Neonates were cultured for 48 h. Preparation of during the experiments and no organic additive was used. homogenates and the determinations of AChE activity and For conventional acute tests, 20 animals were used per protein content were performed as described above for the treatment, in groups of 5 per 100 ml of test solution. Each in vitro AChE inhibition tests. TOXICITY OF MOLYBDENUM AND CHROMIUM TO Daphnia magna 255

TABLE 1 TABLE 2

Range of Concentrations of Molybdenum and Chromium 48-h LC50 Values of Metallic Elements (Chromium and Tested in Conventional Acute, Chronic, in Vivo, and in Vitro Molybdenum) for Daphnia magna and Their Corresponding AChE Inhibition Tests 95% Con5dence Limits

Chromium Molybdenum \ Chemical (mg L ) 48-h LC 95% CL (mg L\) (mg L\) Chromium 0.29 0.269}0.315 Conventional acute tests 0.075}0.6 1500}48,000 Molybdenum 2847.5 2838.7}2857.0 In vivo AChE Inhibition tests 0.025}0.3 250}3000 In vitro AChE Inhibition tests 12.5}100 12.5}100 21-day chronic tests 0.0125}0.125 50}250 3. RESULTS Note. Values are total concentrations of each metallic element. In both acute and chronic tests, oxygen levels were always 2.4. Chemicals and Test Solutions above 7.0 mg L\ and pH variation was always lower than Acetylthiocholine, acid dithiobisnitrobenzoate, and c-bo- 1U. vine globulins were purchased from Sigma. Bradford re- agent was from Bio-Rad, and all the other chemicals were 3.1. Acute Tests from Merck. For all the tests, stock solutions were prepared in nanopure water (conductivity (5 lS/cm). The range of The 48-h LC values determined for each metal element concentrations tested for each chemical is summarized in with the correspondent 95% con"dence limits are presented Table 1. in Table 2.

2.5. Data Analysis 3.2. Chronic Tests

LC values for conventional acute tests and EC values Each test organism was checked for body length at birth for in vivo AChE inhibition assays were determined by and at the end of the test, as well as for the total number of probit analysis (Finney, 1971). Data from chronic tests o!spring (total reproduction). Signi"cant e!ects in total (growth and reproduction) were analyzed with one-way growth and reproduction were found between the di!erent ANOVA. NOEC and LOEC were determined by Dunnett's chromium concentrations and the control (total growth: test (Zar, 1996). Only the organisms for which death occur- F"74.48; df"6,63; P(0.05) (reproduction: F"174.68; red after delivery of the "rst brood and before the end of the df"6,63; P(0.05). For both reproduction and growth, \ \ tests were considered for statistical analysis. EC values for LOEC was 0.0125 mg L and NOEC was 0.025 mg L reproduction and growth were also calculated using probit (Table 3). For mortality, LOEC was 0.01 mg L\ and analysis (Finney, 1971). NOEC and LOEC values for mor- NOEC 0.075 mg L\ (Fisher Exact test: A"10; B"10; tality were determined using Fisher's Exact test (EPA, 1989) a"10; b"6; P"0.043). and 21-day LC values were calculated by probit analysis Relative to total growth, signi"cant e!ects were also ob- (Finney, 1971). Data from AChE inhibition tests were ana- served between all the molybdenum concentrations and the lyzed using a hierarchical (nested) analysis of variance. control (total growth: F"35.52; df"5,52; P(0.05) (re- NOEC and LOEC were determined by Dunnett's test (Zar, production: F"26.6; df"5,54; P(0.05). For reproduc- 1996). The signi"cance level was 0.05. tion and growth, the LOEC was 50 mg L\ and NOEC was

TABLE 3 Sublethal (Total Growth and Reproduction) and Lethal Toxicity Endpoints Determined in Chronic Tests for Chromium and Molybdenum (Total Concentrations of Metallic Elements)

Total growth Reproduction Mortality Chemical \ (mg L ) NOEC LOEC EC NOEC LOEC EC NOEC LOEC EC

Chromium 0.0125 0.025 0.233 0.0125 0.025 0.047 0.075 0.01 0.524 (0.232}0.234) (0.042}0.053) (0.405}0.677) Molybdenum 50 75 204.1 50 75 102.1 75 100 255.1 (203.7}204.2) (97.1}107.3) (255.0}256.2)

Note. 95% con"dence limits are in parentheses. 256 DIAMANTINO ET AL.

FIG. 3. In vivo e!ects of chromium on AChE activity. Data are ex- FIG. 1. Columns with points indicate the results of AChE activity pressed as the mean of three samples (three measurements per without subtraction of the e!ect due to the color of chromium solution. sample)$SEM. (* signi"cantly di!erent at P40.05). White columns presents the in vitro e!ect of chromium on AChE activity with the subtraction of the e!ect of the compound color. Data are expressed as the mean of three samples (three measurements per sample)$ SEM. (* signi"cantly di!erent at P40.05). 3.4. In Vivo AChE Inhibition Tests Chromium signi"cantly inhibited AChE activity in vivo 75 mg L\ (Table 3). Signi"cant di!erences were also found (Fig. 3) (F"10.97; df"4, 10; P(0.05) with a LOEC value in mortality (Fisher Exact test: A"10; B"10; a"10; of 0.15 mg L\ and a NOEC value of 0.075 mg L\. Mol- b"6; P"0.043; LOEC"100 mg L\, NOEC" ybdenum, in in vivo conditions, signi"cantly inhibited AChE 75 mg L\. activity (Fig. 4) (F"12.76; df"4, 10; P(0.05). The NOEC was 750 mg L\ and the LOEC, 1500 mg L\. EC values for metallic elements determined by the in 3.3. In Vitro AChE Inhibition Tests vivo AChE inhibition tests are provided in Table 4. The results obtained in the in vitro AChE tests suggest that chromium signi"cantly increases the activity of the 4. DISCUSSION enzyme (F"16.15; df"5, 12; P(0.05). However, the color of chromium solutions indicated interferences with the de- Considerable di!erences were found between the toxi- termination of both AChE activity and protein content cities of chromium and molybdenum compounds to (Fig. 1). After removal of the color e!ect, results indicate that D. magna. Sodium molybdate revealed a lower toxicity for chromium did not induce signi"cant alterations in AChE D. magna in all tests performed. The results of acute tests for activity. No signi"cant e!ects of molybdenum on AChE chromium element obtained in this work are similar to " " ' were found (F 3.89; df 5,12; P 0.05) (Fig. 2). corresponding 48-h LC reported in the literature by other

FIG. 2. In vitro e!ects of molybdenum on AChE activity. Data are expressed as the mean of three samples (three measurements per sample)$SEM. TOXICITY OF MOLYBDENUM AND CHROMIUM TO Daphnia magna 257

TABLE 4 ؊1 NOEC, LOEC, and EC50 Values (mg L ) (Concentrations of Metallic Elements) Determined from in Vivo AChE Inhibition Tests

Chemical NOEC LOEC EC Chromium 0.075 0.15 0.632 (0.617}0.647) Molybdenum 750 1500 26,492 (22,258}31,531)

Note. 95% con"dence limits are in parentheses.

FIG. 4. In vivo e!ects of molybdenum on in vivo AChE activity. Data are expressed as the mean of three samples (three measurements per cides. The results obtained here agree with recent studies $ 4 sample) SEM. (* signi"cantly di!erent at P 0.05). indicating that some metals (including chromium) may inhibit cholinesterases under both in vivo and in vitro conditions (Labrot et al., 1996; Payne et al. 1996; Guilhermino et authors (0.42 mg L\) (0.229 mg L\) (Baird et al., 1991; al., 1998). At the concentrations tested in this study, chro- Guilhermino et al., 1997). Molybdenum LC values deter- mium and molybdenum compounds did not inhibit AChE mined in this study are comparable to 96-h LC for sodium in vitro. However, a signi"cant AChE inhibition in vivo was molybdate dihydrate as reported by Sebenik et al. (1990) observed at concentrations under the corresponding LC \ (3940 mg L ) and 48-h LC for &&molybdate'' as reported values. These results suggest that the compounds are bio- by KaH lmaH n (1994) (3220 mg L\). transformed in e!ective inhibitors inside the organism. Fur- The results obtained from chronic tests revealed that both thermore, these results also indicate that the in vivo AChE sodium dichromate and sodium molybdate signi"cantly in- inhibition test is very sensitive, indicating signi"cant e!ects hibited reproduction and growth of D. magna (Figs. 5 and 6). at concentrations lower than those inducing death. The high The e!ects of reproduction observed in chronic tests with sensitivity of the in vivo AChE test toward organophos- chromium are in good agreement with some authors (Anger phates reported by Guilhermino et al. (1996) was expected, et al., 1986). According to them, potassium dichromate considering the mechanism of action of these compounds. concentrations higher than 0.1 mg L\ signi"cantly de- However, the high sensitivity of the in vivo AChE test crease o!spring production and swimming ability of relative to metals found in this study is quite surprising and \ D. magna (EC 0.9 mg L ). Regarding the chronic tests supports the point of view that this test is advantageous for with sodium molybdate, no comparable data could be current ecotoxicity testing relative to conventional acute found in the literature. tests based on mortality. Inhibition of cholinesterases has been widely used as The results obtained in AChE in vitro tests indicate that a biomarker for organophosphate and carbamate insecti- some precautions should be taken regarding the use of

FIG. 5. Total growth (a) and total reproduction (b) for each concentration of chromium and for the control. Data are expressed as the mean$SEM (* signi"cantly di!erent at P40.05). 258 DIAMANTINO ET AL.

FIG. 6. Total growth (a) and total reproduction (b) for each concentration of molybdenum and for the control. Data are expressed as the mean$SEM. (* signi"cantly di!erent at P40.05). in vitro tests with colored compounds, which can lead to an ASTM (1980). Standard Practice for Conducting Acute ¹oxicity ¹ests with over- or underestimation of the e!ects on enzyme activity. Fishes, Macroinvertebrates and Amphibians. Report E-790-80. American for the Testing and Materials, Philadelphia. Baird, D. J., Barber, I., Bradley, M., Calow, P., and Soares, A. M. V. M. 5. CONCLUSION (1989). The Daphnia bioassay: A critique. Hydrobiologia 488/489, 403}406. Molybdenum and chromium compounds are included in Baird, D. J., Barber, I., Bradley, M., Soares, A. M. V. M., and Calow, P. the same list (List II) of European dangerous substances. (1991). A comparative study of genotype sensitivity to acute toxic stress The results of this study indicate great di!erences in toxicity using clones of Daphnia magna Straus. Ecotoxicol. Environ. Saf. 21, 257}265. between sodium dichromate and sodium molybdate to D. Biestek, T., and Weber, J. (1976). Electrolytic and Chemical Conversion magna evaluated in acute, chronic, and acetylcholinesterase Coatings, pp. 1}124. Portcullis Press, Redhill, Warszawa. (Translation inhibition tests, with sodium molybdate being less toxic edited by C. R. Drapee, A. I. M., F. C. S.). than sodium dichromate. Thus sodium molybdate seems to Blaise, C., Weels, P. G., and Lee, K. (1997). Microscale testing in aquatic be preferable to sodium dichromate, at least for some indus- toxicology: Introduction, historial perspective and context. In Micro- trial applications, since it is less toxic and demonstrates scale ¹esting in Aquatic ¹oxicology: Advances, ¹echniques, and Practice relatively good anticorrosive performance. Meanwhile, (P. G. Wells, K. Lee, and C. Blaise, Eds.), pp. 1}9. CRC Press, Boca more studies are being done to complete the evaluation of Raton, FL. toxicity of molybdenum compounds to several organisms. Bradford, M. (1976). A rapid and sensitive method for the quanti"cation of microgram quantities of protein utilising the principle of protein}dye binding. Annals Biochem. 72, 248}254. ACKNOWLEDGMENT CEC*Council of the European Communities (1976). Directive of 4 May 1976 on Pollution Caused by Certain Dangerous Substances Discharged This work was partially supported by FundaiaH o para a Cie( ncia e Tec- into the Aquatic Environment of the Community (76/464/EEC, OJL 129, nologia (PRAXIS XXI-BD/2872/94). 18 May). Cohen, M. D., Kargacin, B., Klein, C. B., and Costa, M. (1993). Mech- ¹ REFERENCES anisms of chromium carcinogenicity and toxicity. Crit. Rev. oxicol. 23(3), 255}281. Ellman, G. L., Courtney, D., Andres, V. Jr., and Featherstone, R. M. (1961). Adema, D. M. (1978). Daphnia magna as a test animal in acute and chronic A new and rapid colorimetric determination of acetylcholinesterase toxicity tests. Hydrobiologia 59, 125}134. activity. Biochem. Pharmacol. 7, 88}95. Almeida, E., Digmantino, T. C., Figueiredo, M. O., and SaH , C. (1998). EPA (1989). Short-¹erm Methods for Estimating the Chronic ¹oxicity of Oxidising alternative species to chromium VI in zinc galvanised steel E/uents and Receiving =aters to Freshwater Organisms. Environmental surface treatment. Part 1-A, Morphological and chemical study. Surface Protection Agency. EPA/600/4/001. Coatings ¹echnol. 106, 8}17. Finney, D. J. (1971). Probit Analysis, 3rd ed. Cambridge Univ. Press, Anger, G., Halstenberg, J., Hochgeschwender, K., Uecker, G., Korallus, U., Cambridge. Knopf, H., Schmidt, P., and Ohlinger, M. (1986). Chromium compounds Guilhermino, L., Barros, P., Silva, M. C., and Soares, A. M. V. M. (1998). In ;llmann1s Encyclopedia of Industrial Chemistry (F. Thomas Campbell, Should the use of inhibition of cholinesterases as a speci"c biomarker for R. Pfe!erkorn, and J. F. Rounsaville, Eds.), Vol. A7, pp. 67}97. Editorial organophosphate and carbamate pesticides be questioned? Biomarkers Advisory Board, Weisheim. 3(2), 157}163. TOXICITY OF MOLYBDENUM AND CHROMIUM TO Daphnia magna 259

Guilhermino, L., Diamantino, T. C., Ribeiro, R., Gonc,alves, F., and Soares, Peakall, D. B. (1994). Biomarkers: The way forward in environmental A. M. V. M. (1997). Suitability of test media containing EDTA for the assessment. ¹oxicol. Environ. News 1, 55}60. evaluation of acute metal toxicity to Daphnia magna Straus. Ecotoxicol. Sebenik, R. F., Burkin, A. R., Laferty, J. M., Leichtfried, G., Meyer, H., Environ. Saf. 38, 292}295. Mitchell, P. C. H., Vuksovich, M. S., Chruch, D. A., Van Riper, G. G., Guilhermino, L., Lopes, M. C., Carvalho, A. P., and Soares, A. M. V. M. Gilliland, J. C., and Thielke, S. A. (1990). Molybdenum and molybdenum (1996). Inhibition of acetylcholinesterase activity as e!ect criterion in compounds. In ;llmann1s Encyclopedia of Industrial Chemistry (B. El- acute tests with juvenile Daphnia magna. Chemosphere 32, 727}738. vers, S. Hawkins, and G. Schhulz, Eds.), Vol. A16, pp. 655}698. Editorial Herbert, A., Guilhermino, L., Da Silva De Assis, H. C., and Hansen, P.-D., Advisory Board, Weisheim. (1995/1996). Acetylcholinesterase activity in aquatic organisms as pollu- Soares, A. M. V. M., Baird, D. J., Barber, I., Bradley, M., and Calow, P. tion biomarker. Zeitschrift Angewandte Zool. 3, 1}15. (1991). Variation in chronic stress-tolerance among arti"cal clone popu- KaH lmaH n, E. (1994). Routes to the development of low toxicity corrosion lations of D. magna Straus. An ecophysiological study.