What is a species?

There is an article in the current Scientific American that tracks the changes in the ways in which species are defined. Of course, the original definitions were based on macro-appearance. These were modified to take into account improved information on evolution, and again modified to take into account our increasing understanding of genomes. The authors point out that the definitions have been extended to organisms that don't reproduce sexually, and seem inadequate to the world of microscopic organisms.

Classifications depend on purpose of the classifier. The Dewey Decimal System is OK if you want to browse in a library by subject, but it won't do much good if you are looking for books of the right color to match your decorating scheme. It may be more useful for a gardener to classify plants by size, color and environmental needs rather than species.

From the point of view of the scientist, it may be quite useful to have a classification scheme that is based on appearance since when classifying a new item for a collection, physical appearance is obvious. On the other hand, given the importance of the theory of evolution in biology, a system of classification that refers to commonalities and divergence in ancestry is obviously useful.

The problem seems to be that the same word, "species", is used to refer to a group of individuals whichever way the group is conceived. "Species" is itself a concept used in the classification of organisms. Darwin thought that the debates over classification were excessive. The set of all individuals can be divided into subgroups by similarity, but there are likely always to be borderline individuals. For most purposes it is most useful to look at the prototypical individuals of a population than the outliers.