Hunt Institute Archives Text Discovery Platform

Page transcription

THE CONSTRUCTION OF A CLASSIFICATION

W. H. Wagner, Jr.

Botanical Gardens
The University of Michigan

How are most classifications actually constructed? In the vast majority of cases, we simply inherit past classifications and modify them, adding here, subtracting there. Classifications were originally constructed on the basis of a few conspicuous characters, and were gradually changed by successive workers. Most active taxonomists are really not much concerned with the theory and philosophy behind the construction of classification; they merely do their work and hope for the best. Most workers who have dealt with the problems have usually had some special "iron in the fire" - expressing phenetic relationships, expressing fossil records, and so on. Nevertheless, the problems of constructing classifications are very serious ones, and of importance to biology. Any value that the present discussion may have lies in its attempt to see the whole picture: What are the problems? What, if anything, can or should be done about them?

We are confronted with a series of dilemmas, and the past dozen or so years have produced a great deal of argument, especially in the pages of such journals as Systematic Zoology and Taxon. What is homology? How can characters be quantified, if at all? Should characters be considered equal in significance, or should they be weighted? We cannot delve into these problems in detail, of course - the problems of actual construction of a classification are by themselves nearly overwhelming.

The rationale or concept which underlies the system determines its ultimate form. Early classifications aimed mainly at identification - either to categorize organisms according to their usages, or for pigeon-holing purposes. A latter-day variation is the goal of purely phenetic classification or assortment only, based upon all available data, all characters counted as equal in importance. Modern "biological" classifications, on the contrary, emphasize evolution. Concepts of primitiveness and specialization play a role, and some characters are more important than others in showing relationships. Even evolutionary taxonomists are not agreed, however, on whether the classification should express levels or lines. This is a special problem for plants, where parallelism and convergence have yielded pteridophytes, gymnosperms and angiosperms. Another question has to do with whether the chronology of the evolutionary branchings should be considered or not in setting up taxonomic categories.

By far the greatest problem in constructing classifications has to do with application of categories. Hierarchial inflation has made many systems unwieldy and difficult to use, where species-groups have been raised to genera, genera to families, families to orders, and so on. The problem of the evolutionary continuum enters here, because much of categorization is based upon gaps in the record. The year-by-year changes in classification systems have discouraged many biological workers, especially those in applied fields, about taxonomy, and the fact is that the instability of taxonomy is its chief drawback.