Commentary The weaker points of fish acute toxicity tests and how tests on embryos can solve some issues Claus Wedekind a,b, * , Beat von Siebenthal a,b , Ruth Gingold b,c a Department of Ecology and Evolution, University of Lausanne, Biophore, 1015 Lausanne, Switzerland b Division of Conservation Biology, University of Bern, 3012 Bern, Switzerland c CICESE, Biological Oceanography Department, 22860 Ensenada, BC, Mexico Received 17 July 2006; received in revised form 18 November 2006; accepted 26 November 2006 We list the main weaknesses of fish acute toxicity tests and suggest that multi-factorial embryo tests could solve many of these issues. Abstract Fish acute toxicity tests play an important role in environmental risk assessment and hazard classification because they allow for first estimates of the relative toxicity of various chemicals in various species. However, such tests need to be carefully interpreted. Here we shortly summarize the main issues which are linked to the genetics and the condition of the test animals, the standardized test situations, the uncertainty about whether a given test species can be seen as representative to a given fish fauna, the often missing knowledge about possible interaction effects, especially with micropathogens, and statistical problems like small sample sizes and, in some cases, pseudoreplication. We suggest that multi-factorial embryo tests on ecologically relevant species solve many of these issues, and we shortly explain how such tests could be done to avoid the weaker points of fish acute toxicity tests. Ó 2006 Elsevier Ltd. All rights reserved. Keywords: Fish acute toxicity test; Ecotoxicology; OECD guidelines; Inferential statistics; Representative sampling; Multi-factorial testing; Pseudoreplication; Embryo test 1. Introduction Synthetic-organic chemicals to be marketed are in many countries required by law to pass a number of tests for envi- ronmental risk assessment and hazard classification. The test procedure depends on the properties of the substance, i.e. its ecotoxicological potential, as well as on the expected quantity to be produced (Newman and Strojan, 1998; Newman, 1995; Fent, 2003b; ECETOC, 2003). As part of such mandatory procedures, juvenile or adult fish are tested in ‘‘fish acute tox- icity tests’’ that are standardized by OECD guidelines (OECD, 1992a). Further tests (e.g. OECD, 1984) are typically based on the results of these first tests. Because acute toxicity tests are used in environmental risk management, it is important to know what kind of inferences are possible with what kind of test designs. The major purpose of the OECD guidelines is to permit a comparison of chemicals with respect to their rel- ative hazard. This requires a maximal standardization of test protocols. However, such a standardization automatically leads to a number of problems when the tests results are used to establish region-specific water quality criteria or to predict the possible influence of a substance on a given fish population. The following list summarizes the main reasons why fish acute toxicity tests only provide first estimates of rel- ative hazards. We also list some more general problems with the existing testing designs. We then recommend testing em- bryos instead of juveniles or adult fish and discuss the main advantages and disadvantages of these tests. * Corresponding author. Department of Ecology and Evolution, University of Lausanne, Biophore, 1015 Lausanne, Switzerland. Tel.: þ41 21 692 42 50; fax: þ41 21 692 42 65. E-mail address: claus.wedekind@unil.ch (C. Wedekind). 0269-7491/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.envpol.2006.11.022 Environmental Pollution 148 (2007) 385e389 www.elsevier.com/locate/envpol