* These paragraphs, on giant tortoises, are extracted from an article that I wrote on a boat called the Beagle (not the real one, which is unfortunately long extinct) in the Galapagos archipelago, and published in the Guardian on 19 February 2005.
* Like the modern ‘theory’ of evolution, it is an established fact in the normal sense of the word: a theory in the first of the OED’s definitions that I quoted in Chapter 1, and renamed ‘theorum’.
* Alas, Holmes never said it (just as Burns never wrote ‘for the sake of’ Auld Lang Syne), but the allusion works because everybody thinks he did.
* Indeed, physical laws of scaling ensure that birds as big as an elephant bird couldn’t indulge in powered, flapping flight at all, no matter how big their wing span. This is because the muscles needed to power such massive wings would need to be so big they couldn’t lift their own weight.
* It is an arresting image: South America and Africa speeding away from each other faster than a man can swim, for forty days continuously.
CHAPTER 10
THE TREE OF COUSINSHIP
BONE TO HIS BONE
What a piece of work is the mammalian skeleton. I don’t mean it is beautiful in itself, although I think it is. I mean the fact that we can talk about ‘the’ mammalian skeleton at all: the fact that such a complicatedly interlocking thing is so gloriously different across the mammals, in all its parts, while simultaneously being so obviously the same thing throughout the mammals. Our own skeleton is familiar enough to need no picture, but look at this skeleton of a bat. Isn’t it fascinating how every bone has its own identifiable counterpart in the human skeleton? Identifiable, because of the order in which they join up to each other. Only the proportions are different. The bat’s hands are hugely enlarged (relative to its total size, of course) but nobody could possibly miss the correspondence between our fingers and those long bones in the wings. The human hand and the bat hand are obviously – no sane person could deny it – two versions of the same thing. The technical term for this kind of sameness is ‘homology’. The bat’s flying wing and our grasping hand are ‘homologous’. The hands of the shared ancestor – and the rest of the skeleton – were taken and pulled, or compressed, part by part, in different directions and by different amounts, along different descendant lineages.
Bat skeleton
The same applies – although with different proportions again – in the wing of a pterodactyl (not a mammal, but the principle still holds, which makes it all the more impressive). This pterodactyl’s wing membrane is largely borne by a single finger, what we would call the ‘little’ finger or ‘pinky’. I confess to a homology-inspired neurosis about so much weight being borne by the fifth finger, because in humans it seems so fragile. Silly, of course, because to a pterodactyl the fifth finger, far from being ‘little’, stretched most of the length of the body, and it presumably would have felt stout and strong, as our arm feels to us. Yet again, it goes to illustrate the point I am making. The fifth finger is modified to bear the wing membrane. All the details have become different, but it is still recognizably the fifth finger because of its spatial relationship to the other bones of the skeleton. This long, stout, wing-supporting strut is ‘homologous’ to our little finger. The word for ‘little finger’ in pterodactylese means ‘ruddy great strut’.
In addition to the true fliers – birds, bats, pterosaurs and insects – lots of other animals glide: a habit that might tell us something about the origins of true flight. They have gliding membranes, which need skeletal support; but it doesn’t have to come from the finger bones as it does in the wings of bats and pterosaurs. Flying squirrels (two independent groups of rodents), and flying phalangers (Australian marsupials, looking almost exactly like flying squirrels but not closely related) stretch a membrane of skin between the arms and the legs. Individual fingers are not required to bear much load, and they are not enlarged. I, with my little-finger neurosis, would be happier as a flying squirrel than as a pterodactyl, because it feels ‘right’ to use whole arms and whole legs to do load-bearing work.
Pterodactyl skeleton
Overleaf is the skeleton of a so-called flying lizard, another elegant forest glider. You can immediately see that it is the ribs, rather than the fingers, or the arms and legs, that have become modified to bear the ‘wings’ – the flight membranes. Once again, the resemblance of the skeleton as a whole to other vertebrate skeletons is completely clear. You could go through every bone, one by one, identifying, in each case, the precise bone to which it corresponds in the human or bat or pterosaur skeleton.
‘Flying lizard’ skeleton
The colugo, or so-called ‘flying lemur’, of the south-east Asian forests resembles the flying squirrels and flying phalangers, except that the tail, as well as the arms and legs, is included in the support structure of the flight membrane. That doesn’t feel right to me, because I can’t imagine what it is like to have a tail at all, although we humans, along with all the other ‘tail-less’ apes, have a vestigial tail, the coccyx, buried beneath the skin. Almost tail-less as we apes are, it is hard for us to imagine what it must be like to be a spider monkey, whose tail dominates the entire spinal column. You can see from the picture on colour page 26 how much longer it is even than the already long arms and legs. As in many New World monkeys (indeed, many New World mammals generally, which is a curious fact, hard to interpret), the spider monkey’s tail is ‘prehensile’, meaning that it is modified for grasping, and it almost seems to end in an extra hand, although it is not homologous to a real hand, and has no fingers. Indeed, the spider monkey’s tail looks very much like an extra leg or arm.
I probably don’t need to spell out the message again. The underlying skeleton is the same as in the tail of any other mammal, but modified to do a different job. Well, the tail itself is not quite the same: the spider monkey tail has an extra allowance of vertebrae, but the vertebrae themselves are recognizably the same kind of thing as the vertebrae in any other tail, including our own coccyx. Can you imagine what it would be like to be a monkey with five grasping ‘hands’ – one at the end of each leg as well as at the end of each arm, and a tail – from any of which you could happily hang? I can’t. But I know that the tail of a spider monkey is homologous to my coccyx, just as the enormously long and strong wing bone of a pterodactyl is homologous to my little finger.
Here’s another surprising fact. A horse’s hoof is homologous to the fingernail of your middle finger (or the toenail of your middle toe). Horses walk on tiptoe, literally, unlike us when we walk on what we call tiptoe. They have almost entirely lost their other toes and fingers. In a horse, the homologues of our index finger and our ring finger, and their hind-leg equivalents, survive as tiny ‘splint’ bones, joined to the ‘cannon’ bone, and not visible outside the skin. The cannon bone is homologous to our middle metacarpal which is buried in our hand (or metatarsal, buried in our foot). The entire weight of the horse – very substantial in the case of a Shire or a Clydesdale – is borne on the middle fingers and middle toes. The homologies, for example to our middle fingers or those of a bat, are completely clear. Nobody could doubt them; and, as if to ram the point home, freak horses are sometimes born with three toes on each leg, the middle one serving as a normal ‘foot’, the two side ones having miniature hooves (see picture overleaf).
Can you see how beautiful it is, this idea of near-indefinite modification over immensities of time, each modified form retaining unmistakable traces of the original? I glory in the litopterns, extinct South American herbivores, not closely related to any modern animals, and very different from horses – except that they had almost identical legs and hooves. Horses (in North America*) and litopterns (in South America, which was in those days a gigantic island, the Panama isthmus being way in the future) independently evolved exactly the same reduction of all the fingers and toes except the middle ones, and sprouted identical hooves on the ends of those. Presumably there aren’t all that many ways for a herbivorous mamma
l to become a fast runner. Horses and litopterns hit upon the same way – reducing all digits except the middle one – and they carried it to the same conclusion. Cows and antelopes hit upon another solution: reducing all but two digits.
Polydactylic horse
The following statement sounds paradoxical but you can see how it makes sense, and also how important it is as an observation. The skeletons of all mammals are identical, but their individual bones are different. The resolution of the paradox lies in my calculated usage of ‘skeleton’ for the assemblage of bones, in ordered attachment one to the other. The shapes of individual bones are not, on this view, properties of the ‘skeleton’ at all. ‘Skeleton’, in this special sense, ignores the shapes of individual bones, and is concerned only with the order in which they join up: ‘bone to his bone’ in the words of Ezekiel, and, more vividly, in the song that is based upon the passage:
Your toe bone connected to your foot bone,
Your foot bone connected to your ankle bone,
Your ankle bone connected to your leg bone,
Your leg bone connected to your knee bone,
Your knee bone connected to your thigh bone,
Your thigh bone connected to your hip bone,
Your hip bone connected to your back bone,
Your back bone connected to your shoulder bone,
Your shoulder bone connected to your neck bone,
Your neck bone connected to your head bone,
I hear the word of the Lord!
The point is that this song could apply to literally any mammal, indeed any land vertebrate, and in far more detail than these words suggest. For example your ‘head bone’, or skull, contains twenty-eight bones, mostly joined together in rigid ‘sutures’, but with one major moving bone (the lower jaw*). And the wonderful thing is that, give or take the odd bone here and there, the same set of twenty-eight bones, which can clearly be labelled with the same names, is found across all the mammals.
Human skull
Horse skull
Your neck bone connected to your occipital bone
Your occiput connected to your parietal bone
Your parietal connected to your frontal bone
Your frontal bone connected to your nasal bone
. . .
Your 27th bone connected to your 28th bone . . .
All this is the same, regardless of the fact that the shapes of the particular bones are radically different across the mammals.
Giraffe
Okapi
What do we conclude from all this? We have here confined ourselves to modern animals, so we are not seeing evolution in action. We are the detectives, come late to the scene. And the pattern of resemblances among the skeletons of modern animals is exactly the pattern we should expect if they are all descended from a common ancestor, some of them more recently than others. The ancestral skeleton has been gradually modified down the ages. Some pairs of animals, for example giraffes and okapis, share a recent ancestor. It is not strictly correct to describe a giraffe as a vertically stretched okapi, for both are modern animals. But it would be a good guess (supported by fossil evidence, as it happens, but we aren’t talking about fossils in this chapter) that the shared ancestor probably looked more like the okapi than the giraffe. Similarly, impalas and gnus* are close cousins of each other, and slightly more distant cousins of giraffes and okapis. All four of them are more distant cousins still of other cloven-hoofed animals, such as pigs and warthogs (which are cousins of each other and of peccaries). All the cloven-hoofed animals are more distant cousins of horses and zebras (which don’t have cloven hooves and are close cousins of each other). We can go on as long as we like, bracketing pairs of cousins into groups, and groups of groups of cousins, and (groups of (groups of (groups of cousins))). I have slipped into using brackets automatically, and you know just what they signify. The meaning of the brackets in the following is immediately clear to you, because you already know all about cousins sharing grandparents, and second cousins sharing great-grandparents, and so on:
{(wolf fox)(lion leopard)}{(giraffe okapi) (impala gnu)}
Everything points to a simple branching tree of ancestry – a family tree.
I have implied that the tree of resemblances is really a family tree, but are we forced to this conclusion? Are there any alternative interpretations? Well, just barely! The hierarchical pattern of resemblances was spotted by creationists in pre-Darwinian times, and they did have a non-evolutionary explanation – an embarrassingly far-fetched one. Patterns of resemblance, in their view, reflected themes in the mind of the designer. He had various ideas for how to make animals. His thoughts ran along a mammal theme, and, independently, they ran along an insect theme. Within the mammal theme, the designer’s ideas were neatly and hierarchically bisected into sub-themes (say, the cloven-hoofed theme) and sub-sub-themes (say, the pig theme). There is a strong element of special pleading and wishful thinking about this, and nowadays creationists seldom resort to it. Indeed, as with the evidence from geographical distribution, which we discussed in the last chapter, they rarely discuss comparative evidence at all, preferring to stick to fossils, where they have been taught (wrongly) to think they are on promising ground.
NO BORROWING
To emphasize how odd the idea of a creator sticking rigidly to ‘themes’ is, reflect that any sensible human designer is quite happy to borrow an idea from one of his inventions, if it would benefit another. Maybe there is a ‘theme’ of aircraft design, which is separate from the ‘theme’ of train design. But a component of a plane, say an improved design for the reading lights above the seats, might as well be borrowed for use in trains. Why should it not, if it serves the same purpose in both? When motor cars were first invented, the name ‘horseless carriage’ tells us where some of the inspiration came from. But horse-drawn vehicles don’t need steering wheels – you use reins to steer horses – so the steering wheel had to have another source. I don’t know where it came from, but I suspect that it was borrowed from a completely different technology, that of the boat. Before being superseded by the steering wheel, which was introduced around the end of the nineteenth century, the original steering device of the car was the tiller, also borrowed from boats, but moved from the rear to the front of the vehicle.
If feathers are a good idea within the bird ‘theme’, such that every single bird, without exception, has them whether it flies or not, why do literally no mammals have them? Why would the designer not borrow that ingenious invention, the feather, for at least one bat? The evolutionist’s answer is clear. All birds have inherited their feathers from their shared ancestor, which had feathers. No mammal is descended from that ancestor. It’s as simple as that.* The tree of resemblances is a family tree. It is the same kind of story for every branch and every sub-branch and every sub-sub-branch of the tree of life.
Now we come to an interesting point. There are plenty of beautiful examples where it looks, superficially, as though ideas might have been ‘borrowed’ from one part of the tree and grafted on to another, like an apple variety grafted on to a stock. A dolphin, which is a small whale, looks superficially like various kinds of large fish. One of these fish, the dorado (Coryphæna hippuris) is even sometimes called a ‘dolphin’. Dorados and true dolphins have the same streamlined shape, suited to their similar ways of life as fast hunters near the surface of the sea. But their swimming technique, though superficially similar, was not borrowed from one by the other, as you can quickly see if you look at the details. Although both derive their speed mostly from the tail, the dorado, like all fish, moves its tail from side to side. But the true dolphin betrays its mammal history by beating its tail up and down. The side-to-side wave travelling down the ancestral fish backbone has been inherited by lizards and snakes, which could almost be said to ‘swim’ on land. Contrast that with a galloping horse or cheetah. The speed comes from bending of the spine, as it does with fish and snakes; but in mammals the spine bends up and down, not side to
side. It is an interesting question how the transition was made in the ancestry of mammals. Maybe there was an intermediate stage, which hardly bent its spine at all, in either direction, like a frog. On the other hand, crocodiles are capable of galloping (frighteningly fast) as well as using the lizard-like gait more conventional among reptiles. The ancestors of mammals were nothing like crocodiles, but maybe crocodiles show us how an intermediate ancestor might have combined the two gaits.
Anyway, the ancestors of whales and dolphins were fully paid-up land mammals, who surely galloped across the prairies, deserts or tundras with an up-and-down flexion of the spine. And when they returned to the sea, they retained their ancestral up-and-down spinal motion. If snakes ‘swim’ on land, dolphins ‘gallop’ through the sea! Accordingly, the fluke of a dolphin may look superficially like the forked tail of a dorado, but it is set horizontally, whereas the dorado’s tail fins are aligned in the vertical plane. There are numerous other respects in which the dolphin’s history is written all over it, and I shall come to them in the chapter of that title.
There are other examples where the superficial resemblance is so great that it seems quite hard to reject the ‘borrowing’ hypothesis, but a closer inspection shows that we must. Animals can look so alike that you feel they must be related. But it then turns out that the similarities, though impressive, are outnumbered by the differences when you look at the whole body. ‘Pill bugs’ (see over) are familiar little creatures, with lots of legs, who habitually roll up into a protective ball, like armadillos. Indeed, this may be the origin of the Latin name Armadillidium. That is the name of one kind of ‘pill bug’, which is a crustacean, a woodlouse, related to shrimps but living on land – where it betrays its recent aquatic ancestry by breathing with gills, which have to be kept moist. But the point of the story is that there is a completely different kind of ‘pill bug’ which is not a crustacean at all but a millipede. When you see them rolled up, you’d think they were almost identical. Yet one is a modified woodlouse, while the other is a modified (modified in the same direction) millipede. If you unroll them and look carefully, you will immediately see at least one important difference. The pill millipede has two pairs of legs on most segments, the pill woodlouse only one. Isn’t it beautiful, all this endless modification? A more detailed examination will show that, in hundreds of respects, the pill millipede really does resemble a more conventional millipede. The resemblance to a woodlouse is superficial – convergent.