A General Method of Taxonomic Description for a Computed Similarity Measure

ditionally difficult problem, it has been intensified by the rigorous demands made on data format by the majority of current computer techniques. The failure of computer techniques to classify in some instances can be attributed partly to the gulf between the taxonomist's tacit interpretive approach to biological descriptions and the machine's overly objective treatment of these descriptions. Frequently, the format criterion for computer programs requires that the worker describe his material in a manner unnatural to him and his discipline. The taxonomist's subjective impressions, often drawn from years of experience and training, can be slighted by an overly rigorous numerical method.