The end justifies the means

Abstract

Brummitt (2006) has identified the paraphyly-mono phyly controversy as the most important issue under debate in taxonomy today. In essence the argument is about the merits of the evolutionary (Darwinian) versus phylogenetic (cladistic/Hennigian) approaches towards plant classification. Brummitt's response is in support of the letter by Nordal & Stedje (2005) who pointed out, together with 148 co-signatories, that the rejection of paraphyletic groups as taxa in classification construction by the phylogenetic school is not only a logical impos sibility but is causing unnecessary chaos in plant tax onomy. As a field/herbarium taxonomist in the world's richest temperate flora, that of southern Africa, I can only but endorse this observation. To deny this fact, as some phylogeneticists tend to do (e.g., Dias & al., 2005), is to be out of touch with reality. With more than 20,000 plant species in South Africa alone, the practical value of classifications that are optimally stable, informative and predictive is of utmost importance to come to grips with such astounding diversity. Hence I would like to offer some observations in defense of paraphyletic groups from the perspective of an end-user of plant classifications. In this contribution I argue that practical considerations and the needs and expectations of society should provide the primary guidance in a debate that has hitherto focused mainly on academic issues. For the benefit of readers not familiar with the distinction between evolutionary and phylogenetic taxonomy (mainly espoused and debated in the zoological literature), concise information on the two approaches and its relevance to the current debate is provided. Since completion of the present contribution, the two approaches have also been discussed by H?randl (2007). Evolutionary and phylogenetic classification. ? Classification by its very meaning implies an ordering for practical purposes. It also implies the grouping of objects into classes based on shared characters (similarity). From its inception as folk taxonomies, plant classification has been an applied and practical activity. The success that Linnaeus' artificial sexual system enjoyed was largely due to its simplicity and practical advantages in identification. To address the needs of science, industry and broader society, taxonomists over the years have striven to produce general-purpose (multi-purpose) classification systems, not only by purely phenetic means (including taxomet rics), but since the inception of evolutionary theories increasingly by the use of the best available phylogenetic framework for incorporating attributes of plants from as many fields as possible. The aim is to construct evolu tionary classifications allowing us to store and retrieve information where it is known and predict its presence or absence where it is not. In this way plant taxonomy has not only fulfilled its essential role as an integrative and unifying discipline in botany, but has also enhanced the information storage and predictive value of classification systems. For practical plant identification and information storage and retrieval, the advantages of evolutionary clas sifications over phylogenetic ones (see below) are beyond question. Essentially evolutionary classification is based on the evolution of organisms, not just their phylogeny. Both the evolutionary classification and phylogenetic clas sification are genealogical, but the former is a genealogy of groups (classes) and the latter of clades. Mayr & Bock (2002) define evolutionary classification as 'a classifi cation that duly considers both evolutionary processes, the ecological adaptiveness of evolutionary divergence (degree of difference) and the genealogy (phylogeny) of the taxa.' Mayr & Bock refer to the phylogenetic approach as cladification (instead of classification) and define it as 'an ordering system in which branches of a cladogram, or parts of such branches, are arranged with reference to the sequence of the branching points in the cladogram and based on the principle of holophyly'?holophyly which refers to the cladists' definition of monophyly, namely 'pertaining to a branch of the phyletic tree (and the species on this branch) derived from a stem species (with the first apomorphy diagnostic of this branch) and all of its de scendants, no matter how different.' Inpractice, however, the two approaches have much in common, one of the main distinctions being recognition of paraphyletic taxa by evolutionary taxonomists (strictly speaking, paraphyly as a concept does not exist in an evolutionary classifica tion). The methodological differences between the two