Environmental risk assessment of veterinary pharmaceuticals: development of a standard laboratory test with the dung beetle Aphodius constans.

The environmental risk assessment of veterinary pharmaceuticals for dung beetles is required if the substance is an anti-parasiticide for the treatment of pasture animals. However, the demonstration of the environmental safety of those substances for dung fauna is hampered by the fact that no standardized laboratory test system is currently available. Here a test system using the temperate dung beetle species Aphodius (Agrilinus) constans (Scarabaeidae: Aphodiinae) Duftschmidt is described. The survival of first instar larvae of A. constans exposed to a model substance, dimethoate, spiked into formulated (i.e. dried, formulated and re-wetted) or fresh dung was measured over a period of three weeks. Larvae performed better in formulated dung which also proved to be more suitable for mixing-in test substances homogenously. Dimethoate caused significant larval mortality with LC50 values within a range of 1.3-2.8 mg a.s./kg dung (d.w.), depending on the dung type. Based on the data presented here, it is recommended to incorporate this new test system in the risk assessment process for veterinary pharmaceuticals. However, an international ringtest should to be performed beforehand to ensure adequate validation of the method.

Application of the criteria for classification of existing chemicals as dangerous for the environment.

The criteria for classification and labelling of substances as "dangerous for the environment" agreed upon within the European Union (EU) were applied to two sets of existing chemicals. One set (sample A) consisted of 41 randomly selected compounds listed in the European Inventory of Existing Chemical Substances (EINECS). The other set (sample B) comprised 115 substances listed in Annex I of Directive 67/548/EEC which were classified by the EU Working Group on Classification and Labelling of Existing Chemicals. The aquatic toxicity (fish mortality,Daphnia immobilisation, algal growth inhibition), ready biodegradability and n-octanol/water partition coefficient were measured for sample A by one and the same laboratory. For sample B, the available ecotoxicological data originated from many different sources and therefore was rather heterogeneous. In both samples, algal toxicity was the most sensitive effect parameter for most substances. Furthermore, it was found that, classification based on a single aquatic test result differs in many cases from classification based on a complete data set, although a correlation exists between the biological end-points of the aquatic toxicity test systems.

The effect of tributyltin-oxide on earthworms, springtails, and plants in artificial and natural soils.

Chemical bioavailability in Organisation for Economic Co-operation and Development (OECD) artificial soil can contrast with bioavailability in natural soils and produce ecotoxicologic benchmarks that are not representative of species' exposure conditions in the field. Initially, reproduction and growth of earthworm and Collembolan species, and early seedling growth of a dicotyledonous plant species, in nine natural soils (with a wide range of physicochemical properties) and in OECD soil were evaluated. Soils that supported reproduction and growth of the test species were then used to investigate the toxicity of tributyltin-oxide (TBT-O). Natural soils caused greater toxicity of TBT-O to earthworms (EC(50) values varied from 0.5 to 4.7 mg/kg soil dry weight [dw]) compared with toxicity in OECD soil (EC(50) = 13.4 mg/kg dw). Collembolans were less sensitive to TBT-O than earthworms in natural soils, with EC(50) values ranging from 23.4 to 177.8 mg/kg dw. In contrast, the toxicity of TBT-O to collembolans in OECD soil (EC(50) = 104.0 mg/kg dw) was within the range of EC(50) values in natural soils. Phytotoxicity tests revealed even greater difference between the effects in natural soils (EC(50) values ranged from 10.7 to 189.2 mg/kg dw) and in OECD soil (EC(50) = 535.5 mg/kg dw) compared with results of the earthworm tests. Studies also showed that EC(50) values were a more robust end point compared with EC(10) values based on comparisons of coefficients of variation. These results show that toxicity testing should include studies with natural soils in addition to OECD soil to better reflect exposure conditions in the field.