Friday, September 25, 2015

Dick Brummitt - champion of the paraphyletic

Botanist Dick Brummitt passed away on September 18, 2013.  I missed the opportunity to post a tribute to him at the time, and September of last year also flew by, so in the waning days of September 2015, I will at least mark the second anniversary of his passing.  He is someone we should remember.

Who was Dick Brummitt?  He had a long productive career as a plant taxonomist, primarily at the Royal Botanic Gardens at Kew.  He did what taxonomists do.  He went into the field collecting specimens, did generic revisions and worked on floras.  He did a lot of important work in Africa.  He is more widely known, however,  for his passionate and dogged defense of the peculiar beasts referred to as "paraphyletic taxa."   In a decades-long battle, he pitted himself against the advocates of the new phylogenetic taxonomy, who insisted that only monophyletic groups of organisms could be recognized as formal taxa.  I dedicated two earlier posts to this subject "the great botanical butter battle" and  "making the ancestor problem go away," so I will just briefly summarize it here.

A monophyletic taxon, you may recall, consists of a complete branch, or clade, of a phylogenetic tree: a common ancestor and all of its descendants.  In one of his most memorable essays, Brummitt declared himself to be a "bony fish."  This remark stemmed from the new phylogenetic taxonomy of vertebrates, in which amphibians were a subclade of the bony fish, reptiles (amniotes, if you prefer) a subclade of amphibians, and mammals and birds both subclades of reptiles.  "Mammals" and "bony fishes" are no longer equivalent classes as they were in previous classifications.  The new classification reflects better the evolutionary history of organisms, but raises some difficult practical questions.

What really irked Brummitt was the difficulty of naming ancestral groups.    If we wish to refer to just the bony fishes that did not become amphibians, reptiles, etc., what name can we give them? We cannot put them into a formal category, such as a class, phylum, or family  because they are a paraphyletic taxon.  Such a taxon contains a common ancestor (the first bony fish) and some of its descendants (bony fishes that remained bony fishes), but not all of its descendants (amphibians, etc.), so it is not a complete clade.  If we decide that the bony fish clade is a formal "class" then what are the mammals or birds - sub-sub-sub classes?

Brummitt argued that, taken to its logical conclusion, this situation would lead to a collapse of the taxonomic system, because, after all, bony fishes were just a subclade of an earlier group of vertebrates, the early vertebrates were a subclade of a still earlier group, and so on back to the first organisms.  Recognition of the branching clade structure of life is extremely valuable, but it has made the application of formal ranks difficult, inconsistent, and increasingly less useful.  Brummitt was right on that, and most taxonomists now avoid higher level taxa such as phyla, classes etc., referring instead to stem groups, clades, grades, and other references to portions of the phylogenetic tree.  Supporters of the "phylocode," advocate a system for naming clades without trying to stuff them into formal ranks.  This was a de facto recognition of one of the problems of phylogenetic classification perceived by Brummitt and other like-minded taxonomists.
The groups that we recognize as genera have arisen
sequentially from earlier genera.

With respect to lower level taxa, such as genera and species, we're kind of stuck, however.  We need to give organisms (including fossils) names, and there seems to be no good way to do that other than the traditional binomial: the genus plus the specific epithet.  So we need genera, and we need to be able to group organisms into genera even if they were ancestral to other genera.

For example, the very first species of our own genus, Homo, most certainly evolved from members of an earlier genus traditionally known as Australopithecus, making the latter genus paraphyletic. The same is true for every known genus and the genus that preceded it,  A genus that is monophyletic today might in several million years become paraphyletic by giving birth to new genera. Paraphyletic genera are therefore unavoidable.

Phylogenetic taxonomists have tried to avoid recognizing such genera - sometimes lumping paraphyletic genera with the nearest monophyletic genus, sometimes splitting paraphyletic genera into smaller monophyletic units, or by recognizing unavoidable groups of ancestral species (a"stem group" ) as some kind of special category.  All of these diminish the meaning of genera as comprehensible units of biodiversity.

Brummitt and others (see Hörandl  & Stuessy 2010) argued that paraphyletic genera (and other ranked taxa) should be simply recognized and named in our taxonomic system because they are unavoidable, natural units of evolutionary history.  

Literature cited:

Hörandl  & Stuessy. 2010. Paraphyletic groups as natural units of biological classification. Taxon 59 (6): 1641-1653.