This study was conducted at the Laboratory of Environmental Ecotoxicology (LEQA) within the Department of Natural Resources and Environmental Protection (DRNPA) at the Federal University of São Carlos, Research Center for Agricultural Sciences, located at the Araras Campus, São Paulo State, Brazil. The earthworm species selected for ecotoxicological assessments was Eisenia andrei (Lumbricidae). The specimens, purchased from Minhobox™, were housed in 12-liter plastic containers with temperatures controlled between 18 and 22°C and fed cattle manure weekly.
The experimental soil was obtained from the arable layer of an Oxisol, as classified by the United States soil classification system (2014) and equivalent to a Red Latosol (Dystroferric) according to Embrapa terminology (2018). Chosen for being predominant in Brazil, these soils mirror typical local conditions. The soil, taken from a pesticide-free forest area at a depth of 0–20 cm, was sieved through a 2 mm mesh and air-dried. To ensure the removal of earthworm cocoons and other invertebrates, the soil underwent two 48-hour freezing cycles, interspersed with periods at room temperature (Pesaro et al. 2003).
Table 1
Physical and chemical properties of the Oxisol used for ecotoxicological tests on Eisenia andrei
P-resin | O.M. | pH | K | Ca | Mg | H + Al | SB | CEC | BS | TOC |
mg/dm3 | g/dm3 | Ca/CI2 | mmolc/dm3 | % |
13 | 12 | 4.6 | 2.6 | 43 | no8 | 28 | 53.6 | 81. | 66 | 3.3 |
*P-resin determined using ion exchange resins; pH determination using the CaCl2 method; OM.: Organic Matter; SB: Sum of Bases; BS: Base Saturation. TOC: Total Organic Carbon. Source: Laboratory of Soil Chemistry and Fertility at CCA/UFSCar.
In this research, the herbicide Finale® was used, which contains glufosinate-ammonium as its active component. According to the Agrofit environmental classification system (2024), it falls under Category III, indicating that it is a product hazardous to the environment. The concentrations applied in the various tests are outlined in Table 2.
Table 2
Nominal concentrations of Finale® for avoidance, acute toxicity, and chronic toxicity tests on E. andrei, expressed as mg active ingredient (ai) per kg dry soil (mg kg− 1)
Test | Finale® (mg ai kg− 1) |
Acute toxicity | 0, 175, 340, 505, 670 e 835 |
Chronic toxicity | 0; 3.3; 5; 6.7; 8.3; 10 |
Avoidance test | 0; 3.3; 5; 6.7; 8.3; 10 |
To prepare the solutions, the herbicides were diluted in water. The resulting solution was thoroughly mixed with dry soil in autoclave bags to achieve an even distribution. The soil was then stirred to ensure the herbicide was evenly dispersed throughout. These methods conform to the ecotoxicological testing standards recommended by the ISO - International Organization for Standardization (2008, 2012a, 2012b).
Avoidance test
The study was conducted in strict compliance with the ISO 17512-1 standards (ISO 2008), focusing on independent assessments for different concentrations of the herbicide Finale®. A completely randomized design was utilized, testing six varying concentrations of the herbicide as detailed in Table 2, with each configuration replicated four times. To monitor the inherent behavior of earthworms without the impact of the herbicide, dual controls were used, incorporating uncontaminated soil in both sections of each test container.
Rectangular plastic containers, each 26.2 cm long, 17.7 cm wide, and 8.5 cm high, were employed for the tests. These containers were filled to a depth of 4–5 cm with Oxisol, approximately 300 g of dry soil per container (Niemeyer et al. 2018). Each container was divided into halves; one half was filled with control soil and the other with soil treated with the herbicide, separated by a cardboard divider to avoid mixing before the earthworms were introduced.
Ten adult clitellate earthworms, weighing between 250 and 600 mg each, were positioned at the boundary between the two soil types in each container. The containers were covered with perforated lids to facilitate air circulation and placed in a BOD incubator maintained at 20 ± 2°C, under a 12-hour light/dark cycle. No food was provided during the testing period. After 48 hours, the divider was repositioned to isolate the treated from the untreated soil, and the location of the earthworms in each section was documented.
To calculate the avoidance percentage for the test soil or for each specific herbicide concentration, Equation (I) was employed.
Equation I: \(x \left(\%\right) = \left( \frac{{n}_{c} - {n}_{t}}{N} \right) \times 100\)
where: x = avoidance, expressed as percentage; nc = number of earthworms in the control soil (per test container or in the control soil of all replicates); nt = number of earthworms in the test soil (per test container or in the test soil of all replicates); N = total number of earthworms (typically 10 per test container or in the control soil of all replicates).
In the avoidance tests, the no observed effect concentration (NOEC) and lowest observed effect concentration (LOEC) were calculated using Fisher’s exact test (p < 0.05), applying a one-tailed test format. This statistical approach facilitated comparisons between the actual and anticipated distributions of worms, under the assumption that there would be no avoidance behavior (Natal-Da-Luz et al., 2004). Additionally, the effective median concentration that caused avoidance at 48 hours (AC50-48 h) were estimated using the Trimmed Spearman-Karber statistical method (p < 0.05) (ABNT 2011). These calculations were supported by the Trimmed Spearman-Karber Method software version 1.5 (Hamilton et al., 1977).
Acute toxicity test
The acute toxicity evaluations conformed to the ISO 11268-1 guidelines (ISO 2012a) and were structured using a completely randomized design (CRD) featuring six distinct concentrations with four replications each.
Adult E. andrei earthworms, selected based on a weight range of 250 to 600 mg and age exceeding two months, were exposed to varying levels of Finale®. An initial 14-day screening phase determined the concentration range for the definitive test, which included doses of 0, 1, 10, 100, 500, and 1,000 mg kg− 1. The specific concentrations used in the definitive test are listed in Table 2.
The environmental conditions during the acute toxicity test were consistent with those in the avoidance test, with stable temperature and light cycles maintained throughout the 14-day period. The experimental setup involved 1000 mL containers, each topped with a clear, ventilated lid and filled with soil treated with the herbicide or control soil (only distilled water) to a depth of 5 to 6 cm (500 g). At the beginning of the experiment, a mixture of 5g of cattle manure and 5 mL of distilled water was added to each container to feed the earthworms.
After the 14-day exposure period, the earthworms were carefully removed, and their biomass was measured using an analytical balance. The assessment focused on the survival rates and changes in biomass compared to the initial weights of the earthworms, thereby determining the acute toxicity of the herbicide.
Thus, the median lethal concentration (LC50) was calculated using the Trimmed Spearman-Karber statistical method (p < 0.05). The biomass variation data (mg) were subjected to the Shapiro-Wilk normality test (p > 0.05), Bartlett's test for homogeneity (p > 0.05), analysis of variance (ANOVA), and the means were compared using Dunnett's test (p ≤ 0.05).
Chronic toxicity test
Chronic toxicity testing on the species E. andrei was performed according to the standards outlined in ISO 11268-2 (ISO 2012b). The procedure shared similarities with acute toxicity testing but spanned a duration of 56 days. Throughout the first half of this period, earthworms received weekly feedings of 5 grams of bovine manure and were exposed to various sublethal herbicide concentrations, as outlined in Table 2, with each experimental condition replicated five times.
During the initial 28-day period, earthworms were observed, then extracted from the soil for counting and weighing on an analytical balance to determine their survival and growth rates. After extraction, the remaining soil, cocoons, and juvenile worms were undisturbed in their containers for another 28 days. The study concluded with a comprehensive count and evaluation of the cocoons and juvenile earthworms after 56 days, utilizing a water bath for the final assessment.
The analysis of juvenile earthworm counts in the chronic study included the Shapiro-Wilk W test to check for normality (p > 0.05) and Bartlett’s test for homogeneity of variances (p > 0.05). Subsequently, an ANOVA was conducted to analyze the data, with mean values compared using Dunnett’s test (p ≤ 0.05). Additionally, the exponential model was applied through nonlinear regression analysis to determine concentrations that diminished reproduction by 50% relative to the control group (EC50).