“Our study compellingly demonstrates that, with respect to their mitochondrial DNA, individual animal species from diverse groups are commonly not genetically distinct from their close relatives.”

Many scientists are now studying phylogenetic patterns of molecular genetic variation among closely related species. These studies are often carried out under the assumption that all alleles at a particular locus that are sequenced from individuals of a single species will prove to be "monophyletic," that is, that they will form a discrete group ("clade") on a phylogenetic tree and thus prove to be more closely related to each other than to any alleles from other species. However, such studies sometimes find that this assumption of monophyly does not hold, making it difficult to historically interpret the patterns that are observed. Our paper documents for the first time that these non-monophyletic (i.e., the "paraphyletic" and "polyphyletic" patterns of our title) are actually quite common. More specifically, it shows that particular alleles from one species are often more closely related to alleles from a different species than to other alleles from their own species. Our paper also describes each of the multiple mechanisms potentially responsible for non-monophyly. Thus, our publication provides many scientists with confirmation that their findings are not so crazy after all, as well as a guide for trying to understand the underlying causes of these patterns.

Finally, our results are quite relevant to the ongoing DNA barcoding movement that has attracted so much recent attention. The barcoding program hopes to use species-specific mitochondrial DNA sequences as a means of efficiently identifying collected specimens. However, the ability of this approach to consistently identify species rests on the assumption that species are very generally monophyletic in their mitochondrial alleles. Otherwise, a particular specimen’s sequence may lead it to be erroneously identified as a species with which its actual species shares alleles. Our intensive literature survey, however, revealed mitochondrial non-monophyly to occur in 23% of over 2,300 species across diverse taxa. This percentage is surely an underestimate because many reviewed studies included few individuals and thus fewer opportunities to reject monophyly. Our study thus offers a serious cautionary tale for DNA barcoding.

Our primary discovery was our intensively documented finding that individual species commonly include individuals whose mitochondrial alleles are more closely related to alleles from different species than to those of their own species. That is, species-level non-monophyly (= paraphyly and polyphyly) is common. This is in stark contrast to the common assumption that monophyly is the rule and paraphyly and polyphyly are rare and relatively unimportant. By demonstrating the prevalence of this phenomenon and describing the various mechanisms by which it can occur our paper provides a general resource to researchers confronting these patterns.

Our study compellingly demonstrates that, with respect to their mitochondrial DNA, individual animal species from diverse groups are commonly not genetically distinct from their close relatives. This finding has great implications for studying evolutionary relationships among closely related species and for using evolutionary trees of these species to make biological inferences. It demonstrates that the sampling of individuals and species for such studies must be quite intensive for erroneous interpretations to be minimized. Our study also provides a guide to the various reasons why species might not be genetically distinct at the level of the individual gene.

While I was a graduate student in the early 1990s, I conducted a mitochondrial DNA study of the phylogeny of a group of leaf beetles. Upon including multiple specimens of certain species in my analysis, I detected non-monophyletic patterns in two of them (Funk et al, MOL Biol Evol 12:627 1995a; Funk et al, Evol 49:1008 1995b). At that time, I could find only a handful of similar observations in the literature. The same issue surfaced in a different group of leaf beetles that I was studying for my dissertation (Funk, Mol Biol Evol 16:67 1999). I thus became intrigued with the issue of species-level paraphyly/polyphyly. My co-author Kevin Omland also became interested in this phenomenon upon observing similar patterns in bird phylogenies. The two of us talked for many years about studying this issue systematically through a literature review. We finally did so after having serendipitously procrastinated long enough for a huge body of studies to have accumulated that showed these intriguing patterns and supported our contention of their relevance for evolutionary biologists.