Methods of classifying nemerteans: an assessment

Abstract

Phenetic, cladistic and phyletic methods of classifying animals are discussed with particular reference to nemerteans. It is concluded that phenetic (numerical) taxonomy is particularly inapplicable to any group of invertebrates for which well defined character differences are relatively few, whilst both the phenetic and cladistic methods fail through their fundamental assumption that convergent evolution is a rare occurrence. Terrestrial and freshwater nemerteans especially demonstrate convergent evolution in many ways; cladistic classifications proposed for these animals are therefore untenable. Convergence is shown to be a common occurrence in other nemerteans also. It is concluded that because the traditional phyletic approach does not implicitly assume that resemblances between organisms are more likely to be due to common ancestry than to convergence, it is far more likely to reveal true evolutionary relationships between taxa.

Homoplasy, character function, and nemertean systematics

“…Homoplasy appears to be no more common in nemerteans than among plants and other animals in general. There are no phylogenetic studies on heteronemerteans, but the cladistic studies criticized by Moore & Gibson (Gibson & Moore, 1989;Moore & Gibson, 1993) are on hoplonemerteans (included in Table I). Parsimony aims to give phylogenetic hypotheses that are most consistent with available data and in this sense it does not assume that homoplasies are rare (Farris, 1983;Sober, 1983Sober, , 1988.…”

“…Phylogeny must precede discussions about convergent selection, to maintain that we need to know the function of a character before we can assess its taxonomic value and reaching a taxonomic hypothesis is thus to put the cart before the horse. Moore & Gibson (1993) conclude that cladistics is not applicable to nemertean taxonomy since it "rests on the assumption that convergence is a rare event", whereas "it would be dangerously misleading to assume that convergent evolution [is] a rare phenomenon in any nemertean group", while in fact "convergence is immense" among nemerteans. Moore & Gibson consider their approach better since it "allows the recognition of important characters'' and it has the further advantage of having "fewer set rules to constrain us".…”

“…argued that taxa defined by unique combinations of characters will not necessarily be monophyletic and instead urged for classifications based on reconstructed phylogenies. However, the few attempts to reconstruct nemertean phylogeny by cladistic analyses (Sundberg 1989a(Sundberg , b, 1990 have been criticized by Gibson & Moore (1989) and Moore & Gibson (1993) on the grounds that this approach is not applicable to nemerteans. Moore & Gibson's main objection is that the degree of convergent/parallel evolution is so high in nemerteans that it prohibits the use of parsimony.…”

“…[we will from now on refer to parallelisms, convergences, and reversals as homoplasy .] They instead favour an approach "far more likely to reveal true relationships" (Moore & Gibson, 1993) where "taxonomic importance" of characters is evaluated. Knowing the function of a character is an important part in the process of assessing the taxonomic value of a character (Moore & Gibson, 1973;Gibson & Moore, 1976).…”

Enzyme and colour variation in the hoplonemertean Tetrastemma nigrifrons from the Sea of Japan

“…The morphology of many nemertean species described in the nineteenth century has never been reported (Riser, 1993) . Because well defined character differences in nemerteans are usually few and convergent morphological evolution is common, nemertean taxonomy and classification has several problems (Moore & Gibson, 1993 ;Sundberg, 1993) .…”