Segregation is a spatial outcome of spatial processes that therefore needs to be measured spatially. This is the axiom from which local indices of segregation are developed and applied to the local markets within which schools compete. The indices are used to measure patterns of social segregation between London’s state-funded secondary schools, education authorities, types of selecting and non-selecting schools, and, longitudinally, for cohorts of pupils entering the schools in each of the years from 2003 to 2008. The paper finds sizeable differences between apparently competing schools in the proportions of free school meal eligible pupils they recruit, with selective schools especially and also faith schools under-recruiting such pupils. Whilst there is some evidence that differences between schools have decreased over the period, the trend is considered to be an artefact of using free school meals as a measure of disadvantage, a measure that the paper ultimately questions.