On Phylogenies and Analogies
Our recent book Science and Human Origins has caught the attention of a number of evolutionary bloggers. Go here to get the flavor. Even our Facebook page has seen as spike in commenting. But most have chosen to recycle a review by a graduate student named Paul McBride, so as to avoid reading the book itself, and dealing with its arguments.
His posts (six of them and almost as long as our book!) can be found here. He raises a number of issues that we plan to address. To begin I will discuss his critique that deals with phylogeny.
Mr. McBride doesn't like using cars as an example of design versus common descent, even though another evolutionary biologist once used it as an analogy for evolution. The car analogy was a throwaway comment in my piece, not intended as a serious model for anything. But McBride took it seriously, and said that cars, which are obviously designed, would produce a patchwork pattern of relatedness, with no single tree predominating. In contrast, he said, common descent produces a nested hierarchy of relatedness, with a single species tree that is well-supported and does not require arbitrary weighting of characters.
His argument is similar to that at the TalkOrigins website:
... a cladistic analysis of cars will not produce a unique, consistent, well-supported tree that displays nested hierarchies. A cladistic analysis of cars (or, alternatively, a cladistic analysis of imaginary organisms with randomly assigned characters) will of course result in a phylogeny, but there will be a very large number of other phylogenies, many of them with very different topologies, that are as well-supported by the same data. In contrast, a cladistic analysis of organisms or languages will generally result in a well-supported nested hierarchy, without arbitrarily weighting certain characters. Cladistic analysis of a true genealogical process produces one or relatively few phylogenetic trees that are much more well-supported by the data than the other possible trees.Of course, intelligent design does not necessarily rule out common descent, merely unguided common descent. A biological designer could choose to use descent with modification, and direct the mutational events necessary to produce evolutionary change. But the designer could also add completely new features, or modify and insert previously implemented features from unrelated organisms.
So now the question becomes, what does the biological world really look like? Are gene and species trees neat and tidy, arranged in nested hierarchies, or not? Or do we see evidence of multiple trees with different topologies? Consider this statement from the peer-reviewed literature:
Many of the first studies to examine the conflicting signal of different genes have found considerable discordance across gene trees: studies of hominids, pines, cichlids, finches, grasshoppers and fruit flies have all detected genealogical discordance so widespread that no single tree topology predominates. [Internal citations removed for clarity.]Proposed naturalistic causes of incongruent trees include horizontal gene transfer (mainly in bacteria and plants), incomplete lineage sorting, gene duplication, gene loss, gene flow, and of course, natural selection, which plays havoc with coalescence calculations. We would propose an additional cause.
These patchwork similarities in structure and sequence occur at all levels of biology. For example, so-called convergent evolution (where unrelated organisms share a similar trait) has been known at the morphological level for some time. Think of bird and bat wings or octopus and mammalian eyes. But now with increasing data, patchwork distribution of traits and/or convergence of structures or sequences is being found more and more often, even in animals where horizontal gene transfer is very rare. Examples abound.
Listen to this one from the foremost proponent of convergent evolution, Simon Conway Morris, on the subject of the receptor proteins responsible for olfaction in insects. These molecules seem at first to resemble opsins, G-protein coupled receptors (GPCRs) that span the membrane seven times. But on closer inspection they don't. Even Conway Morris seems puzzled:
At first glance, complete with their seven helices spanning the sensory membrane, they [the insect olfactory receptors] look reassuringly like the ever-reliable GPCRs. Except they aren't! Blink twice and then notice that these proteins are back to front so that the amino-terminal is cytoplasmic and the carboxy-terminal extracellular. This is completely opposite to the GPCRs, but surely it represents a trivial difference? On the contrary. Lurking in the insect "nose" is a ligand-gated cation channel that at first sight looks practically identical to a GPCR but is completely unrelated.So here we have an account of two unique ways of accomplishing the same remarkable task (olfactory reception).
Maybe I am a bear of little brain, but is this not all a little peculiar? Why throw away a perfectly acceptable GPCR -- which after all other ecdysozoans such as nematodes use -- and install what is effectively a near perfect mimic? A little trick to keep us on our Darwinian toes? Maybe a clue comes from the choanoflagellates. Central to their life is nitrogen metabolism, but rather oddly, the genes they employ have been recruited from algae. "If it ain't broke, don't fix it," except that Aurora Nedelcu and colleagues  suggest these imports turned out to be a notch better than the incumbent machinery. Spitfire versus Messerschmitt if you like; both superb aircraft, but the former had the edge.
Similarly, choanoflagellates have borrowed their nitrogen metabolism from algae. But even more remarkably, protists from opposite sides of the phylogenetic tree share unusual organelle ultrastructural features, genome structure, and even methods of RNA editing and processing. Are these things the result of selection and convergent evolution, or something else entirely? No one has a satisfactory answer as yet.
Some argue that there are only a few paths open that can produce particular biological functions (more on that later). That might explain patchy distribution, but not common function with different topologies, like the GPCRs. Rather, proteins that have a common function but different topology, or complex biologic problems that are solved independently but arrive at similar outcomes, suggest the reapplication of existing ideas in new contexts.
This, of course, is precisely the kind of thing that designers do, whether for cars or for creatures.