CapitalistImperialistPig

Richard Dawkins' The Ancestor's Tale is the story of life on Earth, recounted in 39 rendezvous with other forms of life, each rendezvous set at the point of divergence of our ancestor and the other. I always intended to write a review, but now, at the end, I really can't. I am more of a viewer than a reviewer. I like a book where I can get into an argument with the author, and one where I can stop and wonder at the marvels shown, and Dawkins' book marvellously fits that description. Instead of a review, I have presented eleven (now twelve) views of The Ancestor's Tale here . It is a quite wonderful book, and I intend absolutely no disrepect to the author when I say that its merit lies more in the tale than the teller - though he tells it very well. He quite explicitly says the same. The author's love and indeed reverence for his subject is very much in evidence. He says a few words at the end about his disdain for the conventional reverence for the supe...

The final chapter of The Ancestor's Tale consider's a question that had occurred to me earlier in my comments: what would happen if we could rewind to some earlier era and let the whole process occur again? Would evolution retrace its steps? Closely? At all? It turns out that Stuart Kauffman had posed just such questions some time ago. Physicists are addicted to this kind of gedanken experiment, but biologists, not so much. There is a pretty sharp divergence of opinion, it seems. Steven Jay Gould, and in a slightly different context, Ernst Mayr could be considered advocates of "radical contingency." In that view, the replay would be utterly different. Simon Conway Morris takes an almost opposite view, with Dawkins taking an intermediate position a bit closer to Morris. Put me down with Morris. The core of the Gould-Mayr argument is that the detailed path of evolution depends on such an intricate set of contingencies that any replay would rapidly diverge from...

The only redoubt of biology that has not yet yielded utterly to Darwin is the problem of the origin of life. I have now reached Canterbury in The Ancestor's Tale and it's pretty clear that even that last fortress is crumbling fast. The line between living and non-living is still a bright one - there is still a yawning gap between chemistry and life. So how do you get from non-living to life? I'm a firm believer in the idea that there are no "valley crossers" in evolution, to use Lee Smolin's phrase, only hill climbers, so the answer (I think) has got to be some variation on "in small steps." Living things distinguish themselves in a number of ways, but the most obvious are metabolism and reproduction, or more precisely, more or less faithful reproduction. Little pine trees grow up to be big pine trees, not cherry trees, and little kittens grow up to be cats instead of dogs. The hereditary quality is the important one, says Dawkins. Without it y...

If one had the resources and time, a nice experiment would be to try removing all animal life above the unicellular level. Then wait for a billion years or so to see what evolved to fill the niche. How about slime molds? These very curious characters spend the early part of their lives as unicellular amoeboid hunters. After a bit of feasting, some will start sounding the chemical trumpet, and they spontaneously assemble into a slug like organism which will crawl around for a while until it finds a suitable spot to stand on its head and turn itself into a miniature mushroom, with the former rear end becoming a mass of spores to be dispersed and start over. How tough would it be for some of them to learn to prosper and diversify in the slug form? Maybe some parasitic plants might develop some motility? Richard Dawkins, in The Ancestor's Tale , has our last common ancestor with the slime molds (and the rest of the amoebazoans) back at co-ancestor 35, between the fungi and the plant...

There Goes the Neighborhood , or what the protochordate said to the other invertebates when it learned that vertebrates had let four copies of the Hox genes move into one genome instead of the traditional single Hox cluster dwelling. Truer words have rarely been more unspoken. Viewed at the species level, evolution can seem random, chaotic, and rapid - at least if you consider 100,000 years rapid. From the genome's eye view, not so much. Genomes tend to be staid, conservative neighborhoods. The pace of change at the genome level is two or three orders of magnitude slower, but it tends to be cumulative. There are no "new" genes, just variations on the old ones. If one of these variations has the right accent, and can fit in with the guys at the club, it might be allowed to hang around. The discovery of the homeobox genes about a quarter of a century ago was perhaps the most important in the history of embryology. They are a sort of Rosetta stone for the whole of developm...

Three paths diverged in the Sea, and we took the one more travelled. Our fellow lobefins, the coelocanths and lungfish split up about 425 million years ago, and we other tetrapods split from the lungfish very slightly later, about 417 mya, according to Richard Dawkins. Only a very few species of lungfish and coelocanths still survive, but, based on the fossil evidence, they appear to have changed very little over that vast time, while the rest of us lobefins have exploded to become amphibians, reptiles, birds and mammals. Dawkins mentions that continuity of form doesn't necessarily imply genetic stasis. In fact, he says, modern coelocanths and lungfish are about as genetically different from each other as they are from us. If this seems as profoundly counterintuitive to you as it does to me, you might want to check the fine print. It seems that the DNA analyzed to reach this conclusion was all mitochondrial DNA. Mitochondrial DNA is special for a variety of reasons, but most i...

While telling the salamander's tale, Richard Dawkins launches into a critique of Essentialism and what he calls the discontinuous mind. I am broadly sympathetic to that critique, but armed with 100 plus pages of very elementary category theory, I have to say that Dawkins view is also naive. Essentialism enters the picture because it forms a key creationist argument that Dawkins wants to demolish. What about that last common ancestor of both cat and dog, asks the lawyer, was it cat or dog? By way of answer, Dawkins tells him about a couple of pairs of ring species, in each case there is an uninhabitable zone surrounded by a habitable zone. At one end of the ring zone there are two distinct but closely related species, but as you go around the ring each grades continuously into the other. Essentialists, as Mayr calls them, or "discontinuous minds" as Dawkins styles them, seem to have a lot of trouble with this notion of gradation. The Abortion debate is a classic exam...

One of the joys of reading a book like Richard Dawkins' The Ancestor's Tale is learning new and unexpected truths, especially the kind of truth that makes the world make greater sense. It's counterintuitive that the hippo is more closely related to the whale than it is to the pig - more closely related to the whale than it is to any other living animal, in fact. It's perhaps slightly less surprising that crocodiles are closer to birds than they are to turtles, but for me, more surprising that people are closer cousins to trout than trout are to sharks. Appearances can be misleading.

For you sociobiology and evolutionary psychology doubters - Arun sent me something on evo psych, but I fear that it went directly to my locus taedeus - a quote from Dawkins. Do you protest that there aren't 'really' any genes for behavior, only genes for the nerves and muscles that make the behavior? You are still wrecked among heathen dreams. Anatomical features have no special status over behavioral ones where 'direct' effects of genes are concerned. The beaver's 30k genes code not just for teeth and tail, but also for the dam and the lake.

(On Reading The Mouse's Tale, from Richard Dawkins The Ancestor's Tale) The number of genes in a man or a mouse is about 30,000. This is not a tiny number, but to a programmer it seems way too small to specify the program for something as complicated as a man (or a mouse). Each gene specifies one protein, and if you think of each gene as specifying one instruction to the cellular "computer," 30k is not many. How many lines of code are there in Microsoft Office? Millions? Is MS Office 30 (or a hundred or a thousand) times as complicated as a mouse? The thought does give a certain resonance to the idea of "bloatware," but even if you allow that maybe Bill's guys don't exactly make a fetish of efficiency, it still looks incongruous. Dawkins says that the path to understanding is to consider each gene as a subroutine, and that the real power of the cell is exercised in the sequence in which these subroutines are called. To a first approximation, man and...

A couple of odd phrases decorate the Chapter on the Cretaceous Catastrophe in Richard Dawkin's book The Ancestor's Tale. The first is "bullet wounds are hot because of the bullets velocity." Now I had never thought about bullet wounds being hot, possibly since I haven't been shot much, but a fast rifle bullet has enough specific energy to generate some heat. The fastest bullets, travelling at about 1200 m/s, or 1.2 km/s, would probably heat themselves to several hundred degrees upon hitting something hard, like armor plate. The second odd phrase is that cosmic projectiles, like the comet or asteroid that ended the Cretaceous, travel "even faster" than a high speed rifle bullet. Well yes, like tens of times as fast. Collision velocities up to 100 km/s or more are plausible, and velocities smaller than 10 km/s are implausible. So a large cosmic projectile produces specific energies (and temperatures) at least hundreds and most likely thousands of ti...

I've been reading Richard Dawkins's book The Ancestor's Tale and one particular message (so far) stands out. Because inheritance occurs in discrete genes and because genes get shuffled, the ancestral path for different bits of DNA is different. At a particular allele, (non-identical) siblings might each be more closely related to separate chimpanzee individuals and the common ancestors of both humans and chimpanzees than they are to each other. Blood types seem to be an example. I find this to profoundly non-intuitive, but quite undeniable. Overall, of course, the siblings share many more genes with each other than with chimps, but that is in some sense an average effect. Because chimps and humans separated at least 6 million years ago, the separate blood types (and their genes) must have persisted simultaneously through all the 400,000 or so generations in between - there hasn't likely been gene flow in the interim. Of the 2^400,000 ancestral slots from six-millio...