Some of the most egregious abuses of the English language take place at the hands of real-estate agents. Quasi-poetic crimes typical in the U.K. include "delightfully presented" or "well-proportioned." These phrases mean nothing, because when you're trying to sell a crappy house at an inflated price, you write to confuse, not to inform. Here, I'd like to share my disappointment at the abuse of English in the pages of (arguably) the world's most prominent science magazine.
A news item in this week's issue of Nature describes the research of molecular biologist Kevin Peterson under the heading "Phylogeny: Rewriting evolution." In it, we are told that "tiny molecules called microRNAs are tearing apart traditional ideas about the animal family tree" and that Peterson's work "changes everything about our understanding of mammal evolution."
The substance of these claims is a not-yet-published study of molecules called microRNAs that changes something -- I guess "everything" -- about the family tree of mammals. By "family tree" I mean a representation of how they share evolutionary history. Such trees typically show a number of species connected by lines to indicate what pairs have evolved from a common ancestor. For example, we'd place a chimpanzee and human adjacent to one another, to the exclusion of (in order of decreasing relatedness) gorilla, gibbon, lemur, whale, opossum, bird, salamander, lungfish, tuna, and shark. You might be amazed at the potential number of trees by which you could connect these species. For example, we could have placed the gorilla next to the shark, the human with the bird, or the tuna with the whale. In fact, it would take well over 13 billion rearrangements to exhaust the potential number of rooted, bifurcating trees for these 12 species. With more species, this number grows tremendously; around 60 it exceeds the number of atoms in the visible universe.
Now what do these astronomical numbers have to do with mammals and microRNAs? Not much, except to try and convey when it really is accurate for a popular article to use the superlative. At the core of this news essay is a respected Dartmouth biologist who has published quite a bit on microRNAs, small molecules that help regulate protein synthesis. Like other kinds of heritable information, microRNAs are valuable indicators of one of the core postulates of evolution: common descent. If true, closely related animals should have similar microRNAs, just as they should have similar features of the adult skeleton, embryonic development, and DNA. Their use to help tease apart persistent questions about the evolutionary tree of life is a welcome and valuable scientific endeavor.
The papers on microRNAs published to date have indeed sparked debate. For example, in the case of turtles, microRNAs were consistent with a suggestion made in 1924 by Robert Broom (among others) that turtles are reptiles with a close evolutionary relationship to lizards. In the case of jawless fish, microRNAs supported the 1874 tree of Ernst Haeckel, which showed the two living groups -- hagfish and lamprey -- as each other's nearest relative, outside of a larger group of jawed fish.
So what "traditional ideas" are being "torn apart" regarding mammals? Like I mentioned above, nothing has yet surfaced in the peer-reviewed literature, so I can't really say for sure (and neither can the reporter who wrote this piece). However, the news item does come with a figure that contrasts the "traditional tree" (actually just a decade old) with "Kevin Peterson's analysis." At first glance they seem quite different: "Traditionally," the elephant shows up at the base, whereas, according to Peterson, the first branch is mouse and rat. In reality, these two trees show the same pattern, differing only in terms of the root position. In other words, pretend you've got a bonsai plant on your dining room table. Now pick it up, turn it around, and replace the old root with one of the branches that was formerly pointing upwards. Has the pattern of bifurcating branches in your plant actually changed? Not at all. And this is essentially the difference between the two trees figured in this news article. The pattern in each is the same; the two differ only in terms of the root position.
The reality is that in reconstructing the vertebrate tree of life, molecular data, including microRNAs, are quite consistent with comparative anatomy and genomic DNA. None of these data support a tree showing mice with sharks, elephants with lampreys, or any other of the astronomical number of possibilities that one might assemble at random for a given set of species. Some aspects of the mammalian tree of life have indeed changed over the years. For example, using only comparative anatomy, no one would have surmised that hippos are closer to whales than they are to pigs. The rooting of the placental mammal tree is one such topic within mammalogy that maybe Peterson's data can help resolve, once they're vetted through peer review. And by the way, his data are not the first to suggest a placental root among rodents, although many investigators (including me) think this is probably not a genuine signal.
Regardless, the exaggerated prose about "rewriting textbooks" might have been justified if a group like mouse-shark really was supported as sharing a close common ancestor. But there is no such result from microRNAs, because they're far more consistent with other sources of data than this news item would have you believe. The real tragedy when science writing resembles real-estate speak is that the chance is lost to convey to the public a tremendous consilience between different bodies of data. The embryology familiar to T. H. Huxley in the 1860s didn't have to lead to vertebrate trees similar to those based on DNA. But it does, and that is a truly "delightful" fact, one that is a direct result of the common ancestry shared by living things as Charles Darwin theorized 150 years ago. Science journalists should try to make this extraordinary discovery clear to the public rather than obscuring it with buckets of hyperbole.