Abstract
Rendering species occurrence data in an analogous fashion to those of relative rank-order species abundance is advocated. A derived ‘index of numerical importance’, integrating data on both frequency of occurrence and abundance of species within assemblages, is proposed. Changes in rank-order species importance distributions and in individual values of this index are shown to identify the position and nature of spatial change in assemblage structure along the upstream gradient in the Knysna estuarine bay, RSA, and to indicate the species most responsible.