Biological Systematics: Principles and Applications (review)

[Errata]

This second edition of Randall T. Schuh's (2000) manual Biological Systematics: Principles and Applications has been revised and expanded by Andrew V. Z. Brower because biological systematics has become a fast evolving field. At first glance, the contents of the manual do not seem to specifically concern human biologists, but the authors illustrate how cladistic methods have helped to address several questions related to human paleontology, environmental adaptation, and linguistic evolution. Therefore Biological Systematics deserves to be read by population geneticists and human ecologists.

In the first chapter, Schuh and Brower propose an overview of biological systematics. In chronological order they start with Linnaean classification, continue to Mayr's evolutionary taxonomy, and finally develop Sokal's numerical taxonomy and cladistic systematics introduced by Hennig. The cladistics approach is clearly Schuh and Brower's preference. Indeed, one has to recognize that Hennig's approach to systematics represents a complete change of paradigms because it establishes that the evolutionary history of species proceeds from sharp descriptions of the characters defining them.

The next two chapters are more ambitious. Schuh and Brower propose exhaustive thoughts on the philosophical thinking on which systematics was built and proceed to a discussion and definition of the terms and concepts related to such a theory (models, hypotheses, inductive and deductive approaches, consistency and congruence, reconstruction and inference, statistics and probability, similitude and homology, etc). Among others philosophers, Schuh and Brower indicate that Karl Popper is the most influential.

The second part of the book is far more technical (and hence less debatable) and classically describes (innovation is quite difficult in this field) the vocabulary and the basic tools needed to perform a cladistic analysis of real data, especially morphological data. The text generally explains how to describe and code the various hypotheses that can be expressed by character change and provides the meaning and use of some standard concepts, such as homology and homoplasy, synapomorphy and plesiomorphy, monophyly and paraphyly, outgroup, a priori and a posteriori, weighting transformations, independence of traits, and missing data.

The most interesting and controversial part of the manual is the affirmative claim that, to assess phylogenetic relationships, parsimony methods are superior to maximum likelihood ones. Although this standpoint is far from universally accepted, well-known arguments supporting such a view are presented at length. Among them is the fact that maximum likelihood adopts the probability of character changes that can be estimated from molecular data (not easily tested) but not from morphological traits. Another limiting factor of the maximum likelihood approach is that it first needs the definition of an evolutionary model that cannot be refuted and whose degree of realism is difficult to assess. Differently, parsimony methods are presented as independent from evolutionary models. Indeed, one can agree that parsimony is a good criterion even if evolution does not exist, but, in this case, parsimonious trees are interpreted in terms of evolution, thus meaning that parsimony provides trees that a posteriori are no longer model-free.

The third section of the book deals with problems of nomenclature and classification, illustrates different evolutionary scenarios, and introduces ecological and conservation issues. The section seems particularly valuable for those readers interested in the historical establishment of classification methods, from Linnaeus to the more recent International Code of Botanical and Zoological Nomenclatures. Schuh and Brower elucidate classification rules and explain why the scientific community is facing a growing discrepancy between the classical hierarchy proposed by the Linnaean system, the cladistic classification that groups species on synapomorphic bases, and the phylogenetic classification based on common ancestry. Even though Schuh and Brower discuss advantages and weaknesses of alternative methods, their preference goes, not unexpectedly, to cladistic classifications.

The final chapters focus on the methods linking the evolution of species, depicted by a tree or a cladogram, and their dispersal in a given geographic area (or the evolution of their parasites). Schuh and Brower show how, by adjusting ecological traits or adaptive functions on a cladistic tree...