The tree, the network, and the species

Abstract

To enrich the Hennigian internodal conception of species, a new formalization of the definition of the species concept is proposed. This rigorous definition allows for considerable unification of the various, and sometimes conflicting, techniques of species delimitation used in practice. First, the domain of such a definition is set out, namely, the set of all organisms on Earth, past, present, and future. Next, the focus is on the genealogical relationship among organisms, which provides the key to analysing the giant or global genealogical network (GGN) connecting all these organisms. This leads to the construction of an algorithm revealing the topological structure of the GGN, from families to lineages, ending up with a definition of species as equivalence classes of organisms corresponding to branches of the 'tree of life'. Such a theoretical definition of the species concept must be accompanied by various recognition criteria to be operational. These criteria are, for example, the ill-named 'biological species concepts', 'phylogenetic species concepts', etc., usually, but wrongly, presented as definitions of the species concept. Besides clarifying this disputed point, the definition in the present study displays the huge diversity of the scales (time-scale and population size) involved in actual species, thus explaining away the classical problems raised by previous attempts at defining the species concept (uniparental reproduction, temporal depth of species, and hybridization). © 2006 The Linnean Society of London.