Birds are the most diverse group of vertebrate tetrapod. There are over 10,000 species ranging from the massive ostrich (Struthio camelus, where exceptional males can reach 150 kilograms (350 pounds)) to the minuscule bee hummingbird (Mellisuga helenae) at 2 grams (that’s 1/20th of an ounce). They live on every continent and almost every habitat. They include some of the most unusual shapes and behaviors of any animal (with the possible exception of the human political candidate).
Birds are cool. They are also a bit of a mess, taxonomically speaking. Ever since 1990s Phylogeny and Classification of Birds scientists are confident that they have individuals in the right orders, but the relationships of those orders is contentious to say the least. Every other year since 1990 a major work has come out explaining what the correct relationship of all the birds really is.
Fortunately, molecular research is getting faster, more accurate, and (most importantly) cheaper. A very recent study by Prum et.al (A comprehensive phylogeny of birds (Aves) using targeted next-generation DNA sequencing.) has released the latest phylogeny. This is interesting in a couple of ways.
First, this study uses an absolutely monstrous data set. A 2014 study used the whole genome from 48 species. Which, if you are keeping track is less than one half of one percent of all bird species. Still, it’s an impressive effort.
The Prum study uses 198 species (slightly less than 2%) not including the two crocodiles used for an outgroup. They specifically looked at groups whose relationship was in doubt and covered every single order of birds. The supplementary data is interesting. I was trying to determine whether they were using 394 gene sequences per bird or for the whole thing, I found the complete list of species used in the analysis and the justification for using that species. If you’re interested, let me know, I’m not typing them up…
Second, this study uses two different analysis systems. The first is the common maximum likelihood (parsimony) analysis. The second, which seems to be coming into more common use is a Bayesian analysis which corrects a few potential issues from the parsimony system.
The important part isn’t how they work, though I have requested a friend write up an explanation. The important part is that two completely different phylogenetic analysis processes resulted in nearly identical trees. And what’s more interesting is that, for both, the time-scale analysis matched closely with known fossils.
As a side note, another recent study on the evolution of the tyrannosauroid dinosaurs used both methods and has the same excellent matching between the two. (Brusatte, 2016)
What does this study tell us about birds?
First is something really interesting. Flightlessness appears multiple times in birds, not once. This is because the palaeognaths are a sister group to all other birds. The palaegnaths include the ratites (ostrich, emu, kiwi, and a couple of extinct species that make ostriches look like pigeons).
Something to keep in mind is that when these analyses are run, the computer and the math doesn’t care what we call the birds or what birds humans think go in which groups. It makes its own groupings as it discovers close relationships among individuals (which are representative of the entire species). The software in this case, developed 5 groupings in Neoaves (the other major group of birds).
The first, and earliest to split away, are the Strisores. These are the nightjars (which you will find all over the internet in the “find the nightjar” pictures), swifts, and hummingbirds. Within this group, the hummingbirds and the tree swifts of genus Hemiprocne are the most recent divergent species. Recent is a relative term as the estimated time of divergence was 55 million years ago.
The fourth clade is also new. Called the Aequorlitornithes, it consists of all the shorebirds, diving birds, and wading birds. Everything from flamingos to pelicans to penguins (the second flightless bird group). Terns and gulls are also here.
The fifth clade is called Inopinaves and is a large group of all the “land birds”. The strange hoatzin is a sister group to all the other landbirds. This groups includes falcons, parrots, all the perching birds, etc. etc. etc.
The report here supports several hypotheses about the evolution of birds.
This phylogeny upholds the hypothesis that the ancestor of the diurnal swifts and hummingbirds evolved from a clade that had been predominantly nocturnal for ,10 million years. Although hummingbirds have acute near-ultraviolet vision29, the effect of extended ancestral nocturnality on the evolution of the visual system in this group of birds is unknown. Our findings also support the emerging pattern that landbirds evolved from a raptorial grade1 . The sister group relationships of hawks to the rest of the landbirds, of owls to the diverse coraciimorph clade, and of seriemas and falcons to the parrots and passerines indicate the persistence of a raptorial ecology among ancestral landbirds. Lastly, the identification of a new, broadly comprehensive waterbird–shorebird clade indicates a striking, and previously unappreciated, level of evolutionary constraint on the ecological diversification of birds that will be exciting to investigate in the future.
I love cladograms and I’d like to show you this one, but it’s just too large. It’s two pages of very small printed group names.
Still, it’s a fascinating study. I honestly don’t think that this study is the absolute last word in avian phylogeny. But it’s very comprehensive and the use of a second statistical method points out some flaws in previous studies.
Brusatte, S. & Carr, T. (2016). The phylogeny and evolutionary history of tyrannosauroid dinosaurs. Scientific Reports 6, 20252.
Prum, R., Berv, J., Dornburg, A., Field, D., Townsend, J., Lemmon, E. & Lemmon, A. (2015). A comprehensive phylogeny of birds (Aves) using targeted next-generation DNA sequencing. Nature 526(7574), 569–73.