Soil classification has many purposes, the main ones being to organize knowledge, to bring out relationships among the objects classified, to facilitate communication, and to provide a framework for soil management practices and soil research. In Australia, there are currently two 'local' classification systems in use, the so-called Great Soil Group system (Stace et al. 1968) and Northcote's Factual Key (Northcote 1979). Recently, however, the spectre of change has arisen. Firstly, Ray Isbell has been developing a new 'National Australian Soil Classification', currently of 'Second Approximation' status (Isbell 1992a). Secondly, there has been renewed interest in the application of Soil Taxonomy to Australian soils, particularly in South Australia. Such interest was clearly shown at the 4th National Soils Conference, held in Adelaide in April this year, where four papers suggesting changes to Soil Taxonomy were presented. These developments appear to have polarised opinion as to whether changes are necessary, which classification system, or systems, are 'best', and the optimum way of introducing such changes.