Our findings are especially relevant to the newly popular approach known as "DNA barcoding." Barcoding uses the mitochondrial COI gene to attempt to identify unknown individuals, and/or identify cryptic species. DNA barcodes will work best if all species have species-typical DNA. We found that 23% of the 2000+ species in our survey do not show "reciprocal monophyly" in mtDNA. Our results suggest that researchers using DNA barcoding need to proceed with caution. At the very least, many individuals per species should be sampled, and information from morphology, geography, and behavior also needs to be incorporated. However, if done properly, barcoding can help determine with more precision the frequency, causes, and consequences of species-level paraphyly.

Most people, including most biologists, assume that each species on the planet will have species-typical DNA—ravens should have raven DNA; mallard ducks, mallard DNA; polar bears, polar bear DNA and so on. However, because most species share recent ancestry with other closely related species, often there has not been enough time for mutations, natural selection, and enough genetic drift to result in evolutionary distinctiveness of their DNA. In addition, hybridization between distinct species is a very common phenomenon in both animals and plants which also results in intermixing of DNA between two or more species. We have documented that nearly one quarter of all animal species do not have distinctive sequences in their maternally inherited mitochondrial DNA.

During my Ph.D. research I became interested in the case of the intermixed mitochondrial DNA of Mallards and Black Ducks (Avise et al, Evol 1990). I began compiling other examples of this "species-level paraphyly" and realized that there were many other examples. Dan Funk and I began to compare notes on this topic several years ago, and we realized that we had encountered very similar patterns of DNA intermixing although we work on very different groups (insects and birds respectively). During my postdoctoral research, I uncovered a particularly dramatic example: the southwestern US Chihuahuan Raven is deeply nested within the genetic diversity of the widespread Holarctic Common Raven (Omland et al, Proc Roy Soc 2000). I continue to be interested in how species-level polyphyly and paraphyly can impact many research fields from basic areas such as phylogenetics to applied areas such as barcoding and wildlife forensics.