The taxonomic significance of the structure of plant proteins: A classical taxonomist’s view

Abstract

Amino acid sequences of proteins such as cytochrcmec are potentially very useful taxonomically at the level of families and orders, because they are precise, relatively stable, and independent of the traditional characters. Their proper use, however, is complex and difficult. Simple comparisons have very limited value; there are more differences between maize and wheat than between maize and some dicotyledons. Computer-generated phylogenetic trees, produced either by various versions of the matrix method or by the ancestral sequence method, avoid some of the problems of comparison by simple inspection, but they cannot securely distinguish parallel from monophyletic changes, nor can they detect all reversions. The necessary assumption of parsimony must sometimes fail, with unknown consequences. Interpretation of the data faces further theoretical problems if the differences turn out to be selective, and perhaps even more serious ones if they do not. Furthermore, there is some reason to suppose that cytochromec of plants is subject to more frequent evolutionary change than that of animals, and more subject to the fixation of parallel and back mutations. The computer-based trees for cytochromec of a limited number of kinds of angiosperms that have so far been produced are out of harmony with all previous phylogenetic systems, and conclusions that have been based on them are seriously discordant with the fossil record. Although no reliance can be placed on conclusions drawn from the limited data now at hand, it may reasonably be hoped that when the sequence data are available for several different proteins for many different taxa, these data will be very useful in helping to resolve questions of relationships of families and orders.