Evolution and classification : the reformation of cladism / Mark Ridley.
- Mark Ridley
- Date:
- [1986]
Licence: Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
Credit: Evolution and classification : the reformation of cladism / Mark Ridley. Source: Wellcome Collection.
38/220 page 26
![Evolution and Classification character might be a broad adaptation in one group, but not in another. The method supposes this to be relatively unimportant. The observational method, therefore, is to discover which characters are recommended by the study of other specimens, and to rely on them. If the characters are found in fossils, the method could theoretically be used even in the absence of any previous classifica¬ tion. Let us suppose that we have scored our specimens for their numbers of tail vertebrae, numbers of neck (cervical) vertebrae, tooth shape, and dental formula. All these characters are preserved in fossils. If some are more constant than others, if the number of neck vertebrae, for instance, was more constant than the number in the tail, that would be evidence they would be more reliable characters in other specimens. In practice the fossils will have been classified already, which assists the test of constancy because the test is strongest for a known sequence of ancestor-descendant populations (that is, the samples are classified in one taxon). The same method is possible without fossils. Instead of fossils, it can use the evidence of other, previously existing taxonomic groups. The fit of a new character to established taxonomic groups measures the character's reliability. If the character fits a group, being constant across the range of environments occupied by the species of that group, then it is probably reliable; if it changes within the group as if it had been repeatedly evolved and lost, it is probably unreliable. The strength of the test depends on how well established the existing groups are. If the groups are fairly certainly evolutionary, rather than convergent, and a new character fits them, that is strong evidence; if the group is newly erected, tentatively defined by only a few characters, and a new character fits it, that is less strong evidence. In practice, increasingly natural groups are developed, which provide increasingly strong tests of the reliability of new characters. The test simply checks whether a new character has the same taxonomic distribution as other characters, for the groups that the new characters are tested against are themselves defined by characters. The coelom may provide an example. In the nineteenth century, the coelom was found to define a natural group of animals. The Polyzoa, however, in their original extension, comprised two groups, the Entoprocta and Ectoprocta. The former possess a true coleom but the latter do not; the Entoprocta arepseudocoelomate. For this reason above all, Hyman (1951, p. 551) insisted on their separation; 'the insuperable difficulty [for their inclusion in one phylum] is the pseudocoelomate nature of the body cavity of the Entoprocta whereas the Ectoprocta are typical coelomate animals'. The coelom was known to be a good character from other animals and therefore could be supposed to be here too. The separation favoured by Hyman, however, is not universally accepted (Nielsen 1971, 1977); and the coelom 'may be polyphyletic' (Clark 1964, p. 216). 26](https://iiif.wellcomecollection.org/image/b18021451_0039.JP2/full/800%2C/0/default.jpg)
