© 2010 19 th World Congress of Soil Science, Soil Solutions for a Changing World 1 – 6 August 2010, Brisbane, Australia. Published on DVD. 89                    Albert L. Juhasz A,B , Euan Smith A,B , John Weber A,B , Dorota Gancarz C , Matthew Rees C , Allan Rofe C , Tim Kuchel C , Lloyd Sansom D and Ravi Naidu A,B A Centre for Environmental Risk Assessment and Remediation, University of South Australia, Mawson Lakes Campus, Adelaide, 5095, SA. Email Albert.Juhasz@unisa.edu.au, B Cooperative Research Centre for Contamination Assessment and Remediation of the Environment, Mawson Lakes, Adelaide, 5095, SA. C Veterinary Services Division, Institute of Medical and Veterinary Science, Gilles Plains, Adelaide, 5086, SA. D Sansom Institute, University of South Australia, City East Campus, Adelaide 5001, SA.  In this study, the bioaccessibility and relative bioavailability of soil8borne contaminants were compared to determine whether a simple rapid, inexpensive in vitro assay may be used to predict  relative bioavailability for human health exposure assessment. Arsenic, cadmium and lead bioaccessibility in contaminated soil was assessed using a variety of  assays (SBRC, IVG, PBET and DIN) incorporating gastric (G) and intestinal phases (I) while  relative bioavailability was determined using mouse or swine assays. When linear regression models were developed in order to determine the suitability of  assays for predicting arsenic, cadmium and lead relative bioavailability, the correlation between bioaccessibility and relative bioavailability varied depending on the methodology used. While arsenic, cadmium and lead relative bioavailability could be accurately predicted using SBRC8G, PBET8I and Rel8 SBRC8I respectively, a single  method was not suitable for predicting relative bioavailability for all three contaminants.   Arsenic, Bioaccessibility, Bioavailability, Cadmium, Lead, Human Health Exposure Assessment.  Incidental ingestion of contaminated soil is a major non8dietary exposure pathway for many inorganic contaminants. In order to more accurately quantify exposure to inorganic contaminants via soil ingestion, determination of contaminant bioavailability is required. It has been established that arsenic (As), cadmium (Cd) and lead (Pb) bioavailability may be less that 100% as a result of mineralogy, the influence of soil properties and contaminant8soil residence time (ageing) (Ruby  . 1996; Rodriguez  . 1999; Basta  . 2001; Juhasz  . 2007a; 2007b). As a result, exposure and therefore risk to human health may be overestimated if a conservative bioavailability approach is adopted (i.e. 100%).  assays using a variety of animal models (e.g. primate, swine, dog, rabbit, rodent) have been used to quantify the relative bioavailability of contaminants in soil (Freeman   1993; Groen  . 1994; Ng   1998; Roberts   2002; Juhasz   2007b). However, given the time and cost requirements, in addition to ethical issues, there is great demand for an appropriate  assay for estimating relative contaminant bioavailability.  assays are simple, rapid and inexpensive and numerous methods have been applied for the determination of contaminant bioaccessibility (Rodriguez  . 1999; DIN 2000; Oomen   2002; Kelley   2002). However, before these assays can act as a surrogate measure for relative bioavailability, correlation between  bioaccessibility and  relative bioavailability is a mandatory prerequisite for regulatory as well as scientific acceptance. This paper discusses the development, assessment and validation of  assays for predicting the  relative bioavailability of soil contaminated with As, Cd and Pb.    Arsenic, Cd and Pb contaminated soils used in this study were collected from regional areas where the soil type, source and contaminant8soil residence time varied. Soils were air dried then sieved and the <250 Gm particle size retained for chemical characterisation and bioaccessibility / relative bioavailability assessment.