In 2004, the Scottish Executive Education Department (SEED) established a project to pilot restorative practices (RPs) in schools in three local authorities (LAs) in Scotland. The pilot project is one strand of the Scottish Executive's range of initiatives to promote Better Behaviour Better Learning (SEED, 2001) in Scottish schools. In parallel with the pilot project, SEED commissioned a team from the Universities of Edinburgh and Glasgow to evaluate the initiative. That evaluation is in its early stages but its methods will take account of the very different ways of implementing RPs across the eighteen schools participating in the evaluation. This paper will relate those differences to the diffuse nature of RPs themselves and to understandings of the processes by which schools manage change. The questions to be addressed here are: what form will RPs take and what factors are likely to impact upon their implementation in Scottish schools? From its beginnings in the criminal justice field the restorative approach has been reactive, centring on repairing the harm done to relationships and people by offending behaviour. The most accepted working definition among practitioners in that field is that of Tony Marshall (1998): Restorative Justice is a process whereby all the parties with a stake in a particular offence come together to resolve collectively how to deal with the aftermath of offence and its implications for the future. However, although formal processes such as conferencing can be used in schools, RPs need not be confined to such specific applications. McCold and Watchel (2004) suggest a continuum from informal to formal, reflecting that, although a formal restorative practice such as a conference might have a dramatic impact, the informal practices such as daily interactions between teacher and pupil will have a cumulative effect. By this account, RPs offer a permeating approach to increasing empathy within the school and/or classroom community and to building a more positive ethos. Many established strategies such as Circle Time and Peer Mediation would thus exemplify a restorative approach. Very different tacks are open to schools and all of them might be termed restorative. This paper will consider the range of possibilities open to schools and will relate these to factors such as organisational structures, ethos and the nature of leadership, as well as to existing approaches to discipline/behaviour management. Finally, experience of implementing RPs in Australia, in England, and in the USA will be examined and issues for the piloting of RPs in Scottish schools identitfied.