Predicting molybdenum toxicity to higher plants: Estimation of toxicity threshold values

Four plant species (oilseed rape, Brassica napus L.; red clover, Trifolium pratense L.; ryegrass, Lolium perenne L.; and tomato, Lycopersicon esculentum L.) were tested on ten soils varying widely in soil properties to assess molybdenum (Mo) toxicity. A larger range (66-fold-609-fold) of added Mo concentrations resulting in 50% inhibition of yield (ED₅₀) was found among soils than among plant species (2-fold-38-fold), which illustrated that the soils differed widely in the expression of Mo toxicity. Toxicity thresholds based on soil solution Mo narrowed the variation among soils compared to thresholds based on added Mo concentrations. We conclude that plant bioavailability of Mo in soil depends on Mo solubility, but this alone did not decrease the variability in observed toxicity enough to be used in risk assessment and that other soil properties influencing Mo toxicity to plants need to be considered.     

Soil pollution assessment and identification of hyperaccumulating plants in chromated copper arsenate (CCA) contaminated sites, Korea

In recent decades, heavy metal contamination in soil adjacent to chromated copper arsenate (CCA) treated wood has received increasing attention. This study was conducted to determine the pollution level (PL) based on the concentrations of Cr, Cu and As in soils and to evaluate the remediative capacity of native plant species grown in the CCA contaminated site, Gangwon Province, Korea. The pollution index (PI), integrated pollution index (IPI), bioaccumulation factors (BAF_shoots and BAF_roots) and translocation factor (TF) were determined to ensure soil contamination and phytoremediation availability. The 19 soil samples from 10 locations possibly contaminated with Cr, Cu and As were collected. The concentrations of Cr, Cu and As in the soil samples ranged from 50.56-94.13 mg kg⁻¹, 27.78-120.83 mg kg⁻¹, and 0.13-9.43 mg kg⁻¹, respectively. Generally, the metal concentrations decreased as the distance between the CCA-treated wood structure and sampling point increased. For investigating phytoremediative capacity, the 19 native plant species were also collected in the same area with soil samples. Our results showed that only one plant species of Iris ensata, which presented the highest accumulations of Cr (1120 mg kg⁻¹) in its shoot, was identified as a hyperaccumulator. Moreover, the relatively higher values of BAF_shoot (3.23-22.10) were observed for Typha orientalis, Iris ensata and Scirpus radicans Schk, suggesting that these plant species might be applicable for selective metal extraction from the soils. For phytostabilization, the 15 plant species with BAF_root values > 1 and TF values < 1 were suitable; however, Typha orientalis was the best for Cr.     

Derivation of aquatic predicted no-effect concentration (PNEC) for 2,4-dichlorophenol: comparing native species data with non-native species data

2,4-Dichlorophenol (2,4-DCP) is known as an important chemical intermediate and an environmental endocrine disruptor. There is no paper dealing with the predicted no-effect concentration (PNEC) of 2,4-DCP, mainly due to shortage of chronic and site-specific toxicity data. In the present study, toxicity data was obtained from the tests using six Chinese native aquatic species. The HC₅ (hazardous concentration for 5% of species) was derived based on the constructed species sensitivity distribution (SSD), which was compared with that derived from literature toxicity data of non-native species. For invertebrates, the survival no-observed effect concentrations (NOECs) were 0.05 and 1.00 mg L⁻¹ for Macrobrachium superbum and Corbicula fluminea, respectively. NOECs based on fishes’ growth were 0.10, 0.20 and 0.40 mg L⁻¹ for Mylopharyngodon piceus, Plagiognathops microlepis and Erythroculter ilishaeformis, respectively. For aquatic plant Soirodela polyrhiza, NOEC based on concentration of chlorophyll was 1.00 mg L⁻¹. A final PNEC calculated using the SSD approach with a 50% certainty based on different taxa ranged between 0.008 and 0.045 mg L⁻¹. There is no significant difference between HC₅ derived from native and that from non-native taxa.     

Critical Limits for Hg(II) in soils, derived from chronic toxicity data

Published chronic toxicity data for Hg(II) added to soils were assembled and evaluated to produce a data set comprising 52 chronic end-points, five each for plants and invertebrates and 42 for microbes. With end-points expressed in terms of added soil Hg(II) contents, Critical Limits were derived from the 5th percentiles of species sensitivity distributions, values of 0.13 μg (g soil)⁻¹ and 3.3 μg (g soil organic matter)⁻¹ being obtained. The latter value exceeds the currently recommended Critical Limit, used to determine Hg(II) Critical Loads in Europe, of 0.5 μg (g soil organic matter)⁻¹. We also applied the WHAM/Model VI chemical speciation model to estimate concentrations of Hg²⁺ in soil solution, and derived an approximate Critical Limit Function (CLF) that includes pH; log [Hg²⁺]_crit = −2.15 pH −17.10. Because they take soil properties into account, the soil organic matter-based limit and the CLF provide the best assessment of toxic threat for different soils. For differing representative soils, each predicts a range of up to 100-fold in the dry weight-based content of mercury that corresponds to the Critical Limit.     

Cu fractions, mobility and bioavailability in soil-wheat system after Cu-enriched livestock manure applications

Fertilization of crops with livestock manure (LM) is a common waste disposal option, but repeated application of LM containing high concentrations of heavy metals such as Cu could lead to crop toxicity and environmental risk. To examine the Cu availability and uptake by wheat in a Mollisol affected by Cu-enriched LM, pot experiments were conducted. LM (376 mg kg⁻¹ Cu originally) was spiked with different concentrations of Cu (0, 100, 200, 400, 600 and 800 mg kg⁻¹ soil, added as CuSO₄) to simulate soil Cu contamination by LM application. The results indicated that Cu was predominately distributed in organic bound fraction, while the most drastic increase was found in reducible fraction. Acid-extractable fraction played a more important role than other fractions in controlling the mobility and bioavailability of Cu. DTPA-extractable Cu may overestimate the Cu bioavailability since DTPA solution could extract soluble and part of stable forms. The application of LM at 1% level significantly decline the Cu mobility, but that at 3% level exhibited the opposite effect.Although the quantities of Cu in wheat was very low compared with the accumulation in soil, Cu concentrations in roots increased evidently from 12 to 533 mg kg⁻¹ and that in aerial parts were in a narrow range from 12.1 to 32.7 mg kg⁻¹, indicating the more sensitivity of roots to the Cu toxicity. The Cu concentrations in grains after 3% manure application did not approach the threshold for Cu toxicity (<20 mg kg⁻¹) even at higher Cu addition rates.     

Evaluation of various tests for the diagnosis of soil contamination by 2,4,5-trichlorophenol (2,4,5-TCP)

Soil response to contamination with 2,4,5-triclorophenol was studied to test the validity of the concept of Generic Reference Levels (GRL), the main criterion used to define soil contamination. Soil samples were artificially contaminated with doses between 0 and 5000 mg kg⁻¹ of 2,4,5-triclorophenol, and analysed by various tests. Where possible, the response of soils to the contaminant was modelled by a sigmoidal dose-response curve in order to estimate the ED₅₀ values. The tests provided different responses, but only microbial biomass-C and dehydrogenase and urease activities demonstrated soil deterioration in response to contamination. The results suggest that the diagnosis of soil contamination has been greatly simplified in the legislation by the provision of a single figure for each compound, and that the GRL concept could perhaps be substituted by measurement of ED₅₀ values, which better reflect the alteration of a soil due to the presence of a xenobiotic substance.     

Use of Co speciation and soil properties to explain variation in Co toxicity to root growth of barley (Hordeum vulgare L.) in different soils

The influence of soil properties on the bioavailability and toxicity of Co to barley (Hordeum vulgare L.) root elongation was investigated. Ten soils varying widely in soil properties were amended with seven doses of CoCl₂. Soil properties greatly influenced the expression of Co toxicity. The effective concentration of added Co causing 50% inhibition (EC₅₀) ranged from 45 to 863 mg kg⁻¹, representing almost 20-fold variation among soils. Furthermore, we investigated Co toxicity in relation to Co concentrations and free Co²⁺ activity in soil solution. The EC₅₀ values showed variation among soils of 17- and 29-fold, based on the Co concentration in soil solution and free Co²⁺ activity, respectively. Single regressions were carried out between Co toxicity threshold values and selected soil properties. Models obtained showed that soil effective cation exchange capacity (eCEC) and exchangeable calcium were the most consistent single predictors of the EC₅₀ values based on soil added Co.     

Effects of the pharmaceuticals gemfibrozil and diclofenac on biomarker expression in the zebra mussel (Dreissena polymorpha) and their comparison with standardised toxicity tests

Pharmaceuticals, including the lipid regulator gemfibrozil and the non-steroidal anti-inflammatory drug diclofenac have been identified in waste water treatment plant effluents and receiving waters throughout the western world. The acute and chronic toxicity of these compounds was assessed for three freshwater species (Daphnia magna,Pseudokirchneriella subcapitata, Lemna minor) using standardised toxicity tests with toxicity found in the non-environmentally relevant mid mg L⁻¹ concentration range. For the acute endpoints (IC₅₀ and EC₅₀) gemfibrozil showed higher toxicity ranging from 29 to 59 mg L⁻¹ (diclofenac 47-67 mg L⁻¹), while diclofenac was more toxic for the chronic D. magna 21 d endpoints ranging from 10 to 56 mg L⁻¹ (gemfibrozil 32-100 mg L⁻¹). These results were compared with the expression of several biomarkers in the zebra mussel (Dreissena polymorpha) 24 and 96 h after exposure by injection to concentrations of 21 and 21,000 μg L⁻¹ corresponding to nominal concentrations of 1 and 1000 μg L⁻¹. Exposure to gemfibrozil and diclofenac at both concentrations significantly increased the level of lipid peroxidation, a biomarker of damage. At the elevated nominal concentration of 1000 μg L⁻¹ the biomarkers of defence glutathione transferase and metallothionein were significantly elevated for gemfibrozil and diclofenac respectively, as was DNA damage after 96 h exposure to gemfibrozil. No evidence of endocrine disruption was observed using the alkali-labile phosphate technique. Results from this suite of biomarkers indicate these compounds can cause significant stress at environmentally relevant concentrations acting primarily through oxidation pathways with significant destabilization of the lysosomal membrane and that biomarker expression is a more sensitive endpoint than standardised toxicity tests.     

Quantitative assessment of the toxic effects of heavy metals on 1,2-dichloroethane biodegradation in co-contaminated soil under aerobic condition

1,2-Dichloroethane (1,2-DCA) is one of the most hazardous pollutant of soil and groundwater, and is produced in excess of 5.44 × 10⁹ kg annually. Owing to their toxicity, persistence and potential for bioaccumulation, there is a growing interest in technologies for their removal. Heavy metals are known to be toxic to soil microorganisms at high concentrations and can hinder the biodegradation of organic contaminants. In this study, the inhibitory effect of heavy metals, namely; arsenic, cadmium, mercury and lead, on the aerobic biodegradation of 1,2-DCA by autochthonous microorganisms was evaluated in soil microcosm setting. The presence of heavy metals was observed to have a negative impact on the biodegradation of 1,2-DCA in both soil samples tested, with the toxic effect being more pronounced in loam soil, than in clay soil. Generally, 75 ppm As³⁺, 840 ppm Hg²⁺, and 420 ppm Pb²⁺ resulted in 34.24%, 40.64%, and 45.94% increase in the half live (t_½) of 1,2-DCA, respectively, in loam soil, while concentrations above 127.5 ppm Cd²⁺, 840 ppm Hg²⁺ and 420 ppm of Pb²⁺ and less than 75 ppm As³⁺ was required to cause a >10% increase in the t_½ of 1,2-DCA in clay soil. A dose-dependent relationship between degradation rate constant (k₁) of 1,2-DCA and metal ion concentrations was observed for all the heavy metals tested, except for Hg²⁺. This study demonstrated that different heavy metals have different impacts on the degree of 1,2-DCA degradation. Results also suggest that the degree of inhibition is metal specific and is also dependent on several factors including; soil type, pH, moisture content and available nutrients.     

Effect of external iron and arsenic species on chelant-enhanced iron bioavailability and arsenic uptake in rice (Oryza sativa L.)

This study was conducted to investigate the effect of external iron status and arsenic species on chelant-enhanced iron bioavailability and arsenic uptake. Rice seedlings (Oryza sativa L.) were used as model plant, and were grown in artificially contaminated sandy soils irrigated with Murashige and Skoog (MS) culture solution. Arsenate uptake in roots and shoots of rice seedlings were affected significantly (p > 0.05) while dimethylarsinic acid (DMAA) was not by the additional iron and chelating ligand treatments. Regardless of iron concentrations in the soil solution, HIDS increased arsenic uptake for roots more than EDTA and EDDS. Chelating ligands and arsenic species also influenced iron uptake in rice roots. Irrespective of arsenic species, HIDS was found to be more effective in the increase of iron bioavailability and uptake in rice roots compared to other chelants. There was a significant positive correlation (r = 0.78, p < 0.05) between arsenate and iron concentrations in the roots of rice seedlings grown with or without additional iron indicating that arsenate inhibit iron uptake. In contrast, there was no correlation between iron and DMAA uptake in roots. Poor correlation between iron and arsenic in shoots indicated that iron uptake in shoots was neither affected by additional iron nor by arsenic species. Compared to the control, chelating ligands increased iron uptake in shoots of rice seedlings significantly (p < 0.05). Regardless of additional iron and arsenic species, iron uptake in rice shoots did not differed among EDTA, EDDS, and HIDS treatments.     

Use of whole-body and subcellular Cu residues of Lumbriculus variegatus to predict waterborne Cu toxicity to both L. variegatus and Chironomus riparius in fresh water

We tested the use of whole-body and subcellular Cu residues (biologically-active (BAM) and inactive compartments (BIM)), of the oligochaete Lumbriculus variegatus to predict Cu toxicity in fresh water. The critical whole-body residue associated with 50% mortality (CBR₅₀) was constant (38.2-55.6 μg g⁻¹ fresh wt.) across water hardness (38-117 mg L⁻¹ as CaCO₃) and exposure times during the chronic exposure. The critical subcellular residue (CSR₅₀) in metal-rich granules (part of BIM) associated with 50% mortality was approximately 5 μg g⁻¹ fresh wt., indicating that Cu bioavailability is correlated with toxicity:subcellular residue is a better predictor of Cu toxicity than whole-body residue. There was a strong correlation between the whole-body residue of L. variegatus (biomonitor) and survival of Chironomus riparius (relatively sensitive species) in a hard water Cu co-exposure. The CBR₅₀ in L. variegatus for predicting mortality of C. riparius was 29.1-45.7 μg g⁻¹ fresh wt., which was consistent within the experimental period; therefore use of Cu residue in an accumulator species to predict bioavailability of Cu to a sensitive species is a promising approach.     

Toxicity and metabolism of copper pyrithione and its degradation product, 2,2'-dipyridyldisulfide in a marine polychaete

We conducted acute toxicity tests and sediment toxicity tests for copper pyrithione (CuPT) and a metal pyrithione degradation product, 2,2′-dipyridyldisulfide [(PS)₂], using a marine polychaete Perinereis nuntia. The acute toxicity tests yielded 14-d LC₅₀ concentrations for CuPT and (PS)₂ of 0.06 mg L⁻¹ and 7.9 mg L⁻¹, respectively. Sediment toxicity tests resulted in 14-d LC₅₀ concentrations for CuPT and (PS)₂ of 1.1 mg kg⁻¹ dry wt. and 14 mg kg⁻¹ dry wt., respectively. In addition to mortality, sediment avoidance behavior and decreases in animal growth rate were observed; growth rate was the most susceptible endpoint in the sediment toxicity tests of both toxicants. Thus, we propose lowest observed effect concentrations of 0.3 mg kg⁻¹ dry wt. and 0.2 mg kg⁻¹ dry wt. for CuPT and (PS)₂, respectively, and no observed effect concentrations of 0.1 mg kg⁻¹ dry wt. for both CuPT and (PS)₂. The difference in the toxicity values between CuPT and (PS)₂ observed in the acute toxicity test was greater than the difference in these values in the sediment toxicity test, and we attribute this to (PS)₂ being more hydrophilic than CuPT. In addition to the toxicity tests, we analyzed conjugation activity of several polychaete enzymes to the toxicants and marked activity of palmitoyl coenzyme-A:biocides acyltransferase and UDP-glucuronosyl transferase was observed.     

Arsenite toxicity and uptake rate of rice (Oryza sativa L.) in vivo

Toxicity threshold of arsenite on intact rice seedlings was determined and arsenite uptake characteristics were investigated using non-toxic concentrations of arsenite.The arsenite toxicity threshold was 2.4 μM arsenite which reduced growth by 10% (EC₁₀). The two highest arsenite levels induced wilting of seedlings and reduced both, transpiration rate and net photosynthetic rate. Arsenic content in plant tissue increased up to 10.7 μM arsenite and then declined with increasing arsenite concentration in the treatment solution. The contents of Si, P, K, and of micronutrients Cu, Fe, Mn and Zn in shoot d.m. were reduced by arsenite levels ≥ 5.3 μM. In the non-toxic range, arsenite uptake rate was linearly related to arsenite concentration.High arsenite levels reduced growth without being taken up which might be due to increasing binding of arsenite to proteins at the outer side of the plasmalemma.     

Assessing the toxicity of sodium chloride to the glochidia of freshwater mussels: implications for salinization of surface waters

Chloride concentrations in surface waters have increased significantly, a rise attributed to road salt use. In Canada, this may be a concern for endangered freshwater mussels, many with ranges limited to southern Ontario, Canada’s most road-dense region. The acute toxicity of NaCl was determined for glochidia, the mussel’s larval stage. The 24 h EC50s of four (including two Canadian endangered) species ranged from 113-1430 mg Cl L⁻¹ (reconstituted water, 100 mg CaCO₃ L⁻¹). To determine how mussels would respond to a chloride pulse, natural river water (hardness 278-322 mg CaCO₃ L⁻¹) was augmented with salt. Lampsilis fasciola glochidia were significantly less sensitive to salt in natural water (EC50s 1265-1559 mg Cl L⁻¹) than in reconstituted water (EC50 285 mg L⁻¹). Chloride data from mussel habitats revealed chloride reaches levels acutely toxic to glochidia (1300 mg L⁻¹). The increased salinization of freshwater could negatively impact freshwater mussels, including numerous species at risk.     

Effects of municipal solid waste compost and mineral fertilizer amendments on soil properties and heavy metals distribution in maize plants (Zea mays L.)

Soil amendments based on crop nutrient requirements are considered a beneficial management practice. A greenhouse experiment with maize seeds (Zea mays L.) was conducted to assess the inputs of metals to agricultural land from soil amendments. Maize seeds were exposed to a municipal solid waste (MSW) compost (50 Mg ha⁻¹) and NPK fertilizer (33 g plant⁻¹) amendments considering N plant requirement until the harvesting stage with the following objectives: (1) determine the accumulation of total and available metals in soil and (2) know the uptake and ability of translocation of metals from roots to different plant parts, and their effect on biomass production. The results showed that MSW compost increased Cu, Pb and Zn in soil, while NPK fertilizer increased Cd and Ni, but decreased Hg concentration in soil. The root system acted as a barrier for Cr, Ni, Pb and Hg, so metal uptake and translocation were lower in aerial plant parts. Biomass production was significantly enhanced in both MSW and NPK fertilizer-amended soils (17%), but also provoked slight increases of metals and their bioavailability in soil. The highest metal concentrations were observed in roots, but there were no significant differences between plants growing in amended soil and the control soil. Important differences were found for aerial plant parts as regards metal accumulation, whereas metal levels in grains were negligible in all the treatments.     

Effects of calcium peroxide on arsenic uptake by celery (Apium graveolens L.) grown in arsenic contaminated soil

The ability of calcium peroxide (CaO₂) to immobilize As of contaminated soil was studied using pot and field experiments. In pot experiment, CaO₂ applied at 2.5 and 5 g kg⁻¹ significantly increased celery shoot weight and decreased shoot As accumulation, which was ascribed to the formation of stable crystalline Fe and Al oxides bound As and the reduction of labile As fractions in the soil. The labile As fractions were pH dependent and it followed a “V” shaped profile with the change of pH. In field experiment, the dose of CaO₂ application at 750 kg ha⁻¹ was optimal and at which the celery was found to produce the highest biomass (63.4 Mg ha⁻¹) and lowest As concentration (0.43 mg kg⁻¹). CaO₂ probably has a promising potential as soil amendment to treat As contaminated soils.     

Responses to ozone pollution of alfalfa exposed to increasing salinity levels

Stomatal closure and biosynthesis of antioxidant molecules are two fundamental components of the physiological machinery that lead to stress adaptation during plant's exposure to salinity. Since high stomatal resistance may also contribute in counteracting O₃ damages, we hypothesized that soil salinization may increase O₃ tolerance of crops. An experiment was performed with alfalfa grown in filtered (AOT40 = 0 in both years) and non-filtered (AOT40 = 9.7 in 2005 and 6.9 ppm h in 2006) open-top chambers. Alfalfa yield was reduced by O₃ (−33%) only in plants irrigated with salt-free water, while the increasing levels of soil salinity until 1.06 dS m⁻¹ reduced both stomatal conductance and plant O₃ uptake, thus linearly reducing O₃ effects on yield. Therefore a reliable flux-based model for assessing the effects of O₃ on crop yield should take into account soil salinity.     

Fungicide sensitivity of Trichoderma spp. from Agaricus bisporus farms in Serbia

Trichoderma species, the causal agents of green mould disease, induce great losses in Agaricus bisporus farms. Fungicides are widely used to control mushroom diseases although green mould control is encumbered with difficulties. The aims of this study were, therefore, to research in vitro toxicity of several commercial fungicides to Trichoderma isolates originating from Serbian and Bosnia-Herzegovina farms, and to evaluate the effects of pH and light on their growth. The majority of isolates demonstrated optimal growth at pH 5.0, and the rest at pH 6.0. A few isolates also grew well at pH 7. The weakest mycelial growth was noted at pH 8.0–9.0. Generally, light had an inhibitory effect on the growth of tested isolates. The isolates showed the highest susceptibility to chlorothalonil and carbendazim (ED₅₀ less than 1 mg L^?1), and were less sensitive to iprodione (ED₅₀ ranged 0.84–6.72 mg L^?1), weakly resistant to thiophanate-methyl (ED₅₀ = 3.75–24.13 mg L^?1), and resistant to trifloxystrobin (ED₅₀ = 10.25–178.23 mg L^?1). Considering the toxicity of fungicides to A. bisporus, carbendazim showed the best selective toxicity (0.02), iprodione and chlorothalonil moderate (0.16), and thiophanate-methyl the lowest (1.24), while trifloxystrobin toxicity to A. bisporus was not tested because of its inefficiency against Trichoderma isolates.     

Enhancing p-cresol extraction from soil

Soil washing is a potential technology for rapid removal of organic hydrocarbons sorbed to soils. In this work, p-cresol desorption with different non-ionic surfactants (Tween 80, Brij 30 and Triton X-100) was compared to cyclodextrine and citrate as solubilizer. A series of batch extraction experiments were conducted at 20 °C using the field soil with different extracting solutions at various concentrations to investigate the removal efficiency and to optimize the concentration of the extractant. The use of the different extracting agents was very selective to p-cresol extraction, minimizing soil organic matter releasing and maintaining the natural pH of the soil. The highest asymptotic values of desorption percentages were obtained for Tween 80 and Brij 30 at 48 h. However, Brij 30 ecotoxicity (EC₅₀ = 0.5 mg L⁻¹) is in the same order of that obtained for p-cresol, being this surfactant clearly ruled out. Liquid to solid ratio of 2.5 mL g⁻¹ presented the best extraction results, while concentrations higher than 1 g L⁻¹ for Tween 80 and Citrate did not produce any significant effect on the desorption efficiency. p-Cresol extraction efficiencies higher than 70% and 60% for Tween 80 and Citrate, respectively.     

The freshwater planarian Polycelis felina as a sensitive species to assess the long-term toxicity of ammonia

Behavioural endpoints are a good link between physiological and ecological effects. However long-term behavioural endpoints are not uniformly studied over all different organism groups. For example behaviour has been scarcely studied in planarians. Unionized ammonia (NH₃) is one of the most widespread pollutants in developed countries, and is known to alter animal behaviour. In this study a long-term (30 d) bioassay was conducted to assess the effect of this pollutant on survival and behavioural activity (e.g. locomotion activity) of the freshwater planarian Polycelis felina. One control and three environmentally-realistic concentrations of unionized ammonia (treatments of 0.02, 0.05, and 0.09 mg N-NH₃ L⁻¹) were used in quintuplicate. The behaviour of planarians was measured after 0, 10, 20 and 30 d of ammonia exposure. Mortality was recorded every 2 d. Unionized ammonia increased mortality in the two highest NH₃ concentrations and the locomotory activity was depressed in all treatments after 20 d of exposure. Behavioural effect was observed at concentrations 20 times lower than the short-term LC50 for this species. Previous studies proposed safe concentrations of unionized ammonia of 0.01-0.10 mg N-NH₃ L⁻¹ to aquatic ecosystems, but our study has shown that these concentrations will affect planarians. Because planarians play a key role in streams (as predator/scavenger), safe concentrations should be below 0.02 mg N-NH₃ L⁻¹ to protect this species in the freshwater community. Our results can contribute to improve the knowledge about ammonia toxicity to freshwater ecosystems, we recommend that safe concentrations of unionized ammonia should be based on very sensitive species.