Behavioural Endpoints in Earthworm Ecotoxicology Evaluation of Different Test Systems in Soil Toxicity Assessment

Background, Aims and Scope. Endpoints in earthworm ecotoxicology scheduled in guidelines are mortality and reproduction rates. However, not only the direct influence of pollutants on population parameters but also changes in behaviour such as substrate avoidance can have an important impact on soil ecosystems. In practice two different avoidance response tests are applied in earthworm ecotoxicology: (i) a six-chamber test system and (ii) a two-chamber test system. Both avoidance response-test systems were compared to establish their respective advantages and disadvantages in order to advance the standardisation of behavioural tests. The earthworm avoidance-response tests were applied in addition to the standard acute and chronic earthworm toxicity tests (ISO 11268) in order to compare the sensitivity of the test endpoints. Methods. Test substrates were contaminated with crude oil and 2,4,6-Trinitrotoluene (TNT), respectively. The test species was Eisenia fetida. The earthworms were exposed to the contaminated substrates and their mortality (14 d), reproduction rates (number of cocoons after 28 d, juvenile hatching after 56 d), and substrate preference (48 h) determined. Results and Discussion. Whereas 1000 mg/kg TPH (Total Petroleum Hydrocarbons) did not show any lethal effects, 100% mortality occurred in soil with comparable TNT concentration. The acute tests consistently produced the highest effect concentrations whereas reproduction and substrate avoidance were the more sensitive test parameters. Both behavioural test systems, when compared, showed similar substrate avoidance after an incubation time of 48 h. The six-chamber test system provides the potential to test six different substrates/concentrations at one time. It was observed, however, that earthworms did not migrate among all test chambers within a test unit in order to select the most appropriate substrate. Orientation was observed only between directly neighbouring test compartments, which complicates the interpretation of the test results. Conclusion. Substrate avoidance and reproduction variables were clearly more sensitive test endpoints than mortality. Therefore avoidance-response tests proved to be useful test methods in detecting effects of sublethal concentrations of pollutants on earthworms. The test duration of the avoidance tests is much shorter compared to the standard acute and chronic earthworm toxicity tests, which makes them a quick screening tool for identifying potential soil toxicity. Both avoidance-response test systems showed comparable results regarding the test sensitivity. Nonetheless, the incomplete substrate use in the six-chamber avoidance test due to the reduced migration possibilities (orientation only to neighbouring chambers) might reduce the distinctness of test results as it allows only reliable information on Introduction Earthworms are common test organisms in terrestrial ecotoxicology. Both an acute earthworm toxicity test and a reproduction test with the endpoints mortality, reproduction (cocoon production and juvenile hatching) and adult body weight change are standardised and well described in guidelines (ISO 11268-1/2). Other endpoints such as behavioural, morphological and physiological changes are reported occasionally, but they have not been evaluated in a standardised way (Kula 1998). Endpoints of the standardised earthworm toxicity tests reflect direct effects (lethal and sublethal) of chemicals, whereas behavioural tests focus on indirect effects. The reduction of population size due to mortality or reduced reproduction is an ecological consequence of exposure to chemicals in soil. However, behavioural changes such as substrate avoidance can be also ecological relevant. Emigration of earthworms and the subsequent loss of their beneficial functions in soil (aeration, drainage, enrichment of organic material, etc.) can lead to a degradation of soil qualities. Additionally, loss of earthworms from an area might also affect the numbers and distribution of their vertebrate predators. Thus, migration of earthworms can impact an ecosystem. Consequently, tests with behavioural endpoints should be included in ecotoxicological test batteries in order to assess the toxic impacts of chemicals on soil ecosystems. While Darwin (1881) showed that earthAvoidance Response Test Research Articles 80 JSS – J Soils & Sediments 3 (2) 2003 worms select between different foods, recent tests proved that earthworms respond to chemical stimuli (Slimak 1997, Mather and Christensen 1998). High numbers of chemoreceptors, concentrated in the prostomium and anterior segments and the distribution of epidermal tubercles and nerve endings in and around singular body segments contribute to the capacity of earthworms to react to chemicals in their environment (Wallwork 1983). The epithelium in the mouth region accommodates groups of sensory cells, which can be stimulated by chemical substances associated with taste. These cells are associated with the selection of food, the detection of unfavourable environmental conditions (e.g. soil acidity) (Mangold 1953), and the detection of mucus secretions of other earthworms (Edwards and Lofty 1972). This sensitivity towards chemicals, coupled with their locomotory abilities enable earthworms to avoid adverse habitats (Stephenson et al. 1998). Avoidance-response tests reflect these behavioural properties of the earthworms. Principle of avoidance response tests is the preference or avoidance of substrates after a specific exposure period. The scope of this study included (i) a comparison of the results of the behavioural tests (six-chamber avoidance response test) with those from the standardised earthworm toxicity tests using soils contaminated with crude oil and 2,4,6-Trinitrotoluene (TNT) and (ii) a comparison of two different earthworm avoidance response tests. The first avoidance test system was developed by Stephenson et al. (1998) and consists of six compartments per test container. The second test system, described by Yeardly et al. (1996) and more precisely by Hund-Rinke and Wiechering (2001), uses test units that are divided into two equal sections. Both avoidance-response test systems were compared to identify the advantages and limitations of the two systems to make a further step into the direction of a standardisation of behavioural tests. 1 Material and Methods