Commentary: do we have a consistent terminology for species diversity? We are on the way.

A consistent terminology for species diversity is subject of an ongoing debate. Recently Tuomisto (Oecologia 164:853-860, 2010) stated that a consistent terminology for diversity already exists. The paper comments on recent papers by ourselves (Jurasinski et al. Oecologia 159:15-26, 2009) and by Moreno and Rodriguez (Oecologia 163:279-282, 2010). Both started from Whittaker's diversity concept to discuss the ambiguities of the terminology and propose a new, more consistent terminology that is based on the different approaches to diversity analysis. In contrast, Tuomisto adheres to a strict school of thinking and derives a diversity framework in the sense of Whittaker (alpha, beta, gamma) from the conceptual definition of diversity itself. A third group of papers discusses appropriate methods for the analysis of the variation in species composition. Here, we support the idea that alpha, beta and gamma diversity should be used in a strict sense that is based only on the conceptual definition of diversity. We accordingly extend and modify our terminological concept for species diversity. All approaches to the analysis and quantification of species composition and diversity can be assigned to three abstraction levels (species composition, variation in species composition,and variation in variation in species composition) and two scale levels (sample scale, aggregation scale). All methods that investigate the variation in species composition across scale levels evaluate beta relation with beta diversity being just one form of beta relation, which is calculated by dividing gamma diversity of order q by the appropriate alpha diversity of the same order. In contrast, differentiation refers to a pairwise calculation of resemblance in species composition. It is restricted to sample scale and is therefore most often only an intermediate step of analysis. Many ecological questions can be addressed either by direct analysis of the variation in species composition using raw data approaches or by further analysis of differentiation datasets on aggregation scale with or without respect to an external gradient.