Endless Forms Most Beautiful And Most Wonderful Have Been And Are Being Evolved

I'm not sure what Darwin meant by 'endless'. It could have been just a superlative, deployed to soup up 'most beautiful' and 'most wonderful'. I expect that was part of it. But I like to think that Darwin meant something more particular by 'endless'. As we look back on the history of life, we see a picture of never-ending, ever-rejuvenating novelty. Individuals die; species, families, orders and even classes go extinct. But the evolutionary process itself seems to pick itself up and resume its recurrent flowering, with undiminished freshness, with unabated youthfulness, as epoch gives way to epoch.

Let me briefly return to the computer models of artificial selection that I described in Chapter 2: the 'safari park' of computer biomorphs, including arthromorphs and the conchomorphs that showed how the great variety of mollusc shells might have evolved. In that chapter, I introduced these computer creatures as an illustration of how artificial selection works and how powerful it is, given enough generations. Now I want to use these computer models for a different purpose.

My overwhelming impression, while staring into the computer screen and breeding biomorphs, whether coloured or black, and when breeding arthromorphs, was that it never became boring. There was a sense of endlessly renewed strangeness. The program never seemed to get 'tired', and nor did the player. This was in contrast to the 'D'Arcy' program that I briefly described in Chapter 10, the one in which the 'genes' tugged mathematically at the coordinates of a virtual rubber sheet on which an animal had been drawn. When doing artificial selection with the D'Arcy program, the player seems, as time goes by, to move further and further away from a reference point where things made sense, out into a no-man's-land of mis-shapen inelegance, where sense seems to decrease the further we travel from the starting point. I have already hinted at the reason for this. In the biomorph, arthromorph and conchomorph programs, we have the computer equivalent of an embryological process - three different embryological processes, all in their different ways biologically plausible. The D'Arcy program, by contrast, doesn't simulate embryology at all. Instead, as I explained in Chapter 10, it manipulates the distortions by which one adult form may be transformed into another adult form. This lack of an embryology deprives it of the 'inventive fertility' that the biomorphs, arthromorphs and conchomorphs display. And the same inventive fertility is displayed by real-life embryologies, which is a minimal reason why evolution generates 'endless forms most beautiful and most wonderful'. But can we go beyond the minimal?

In 1989 I wrote a paper called 'The evolution of evolvability' in which I suggested that not only do animals get better at surviving, as the generations go by: lineages of animals get better at evolving. What does it mean to be 'good at evolving'? What kinds of animals are good at evolving? Insects on land and crustaceans in the sea seem to be champions at diversifying into thousands of species, parcelling up the niches, changing costumes through evolutionary time with frolicsome abandon. Fish, too, show amazing evolutionary fecundity, so do frogs, as well as the more familiar mammals and birds.

What I suggested in my 1989 paper is that evolvability is a property of embryologies. Genes mutate to change an animal's body, but they have to work through processes of embryonic growth. And some embryologies are better than others at throwing up fruitful ranges of genetic variation for natural selection to work upon, and might therefore be better at evolving. 'Might' seems too weak. Isn't it almost obvious that some embryologies must be better than others at evolving, in this sense? I think so. It is less obvious, but nevertheless I think a case can be made, that there might be a kind of higher-level natural selection in favour of 'evolvable embryologies'. As time goes by, embryologies improve their evolvability. If there is a 'higher-level selection' of this kind, it would be rather different from ordinary natural selection, which chooses individuals for their capacity to pass on genes successfully (or, equivalently, chooses genes for their capacity to build successful individuals). The higher-level selection that improves evolvability would be of the kind that the great American evolutionary biologist George C. Williams called 'clade selection'. A clade is a branch of the tree of life, like a species, a genus, an order or a class. We could say that clade selection has occurred when a clade, such as the insects, spreads, diversifies and populates the world more successfully than another clade such as the pogonophora (no, you probably haven't heard of these obscure, worm-like creatures, and there's a reason: they are an unsuccessful clade!). Clade selection doesn't imply that clades have to compete with each other. The insects don't compete, at least not directly, with the pogonophora for food or space or any other resource. But the world is full of insects, and almost devoid of pogonophora, and we are rightly tempted to attribute the success of the insects to some feature that they possess. I am conjecturing that it is something about their embryology that makes them evolvable. In the chapter of Climbing Mount Improbable entitled 'Kaleidoscopic Embryos' I offered various suggestions for specific features that make for evolvability, including constraints of symmetry, and including modular architectures such as a segmented body plan.

Perhaps partly because of its segmentally modular architecture, the arthropod clade* is good at evolving, at throwing up variation in multiple directions, at diversifying, at opportunistically filling niches as they become available. Other clades may be similarly successful because their embryologies are constrained to mirror-image development in various planes.1 The clades that we see peĀ°pling the lands and fte seas are the clades that are good at evolving. In clade selection, unsuccessful clades go extinct, or fail to diversify to meet varying challenges: they wither and perish. Successful clades blossom and flourish as leaves on the phylogenetic tree. Clade selection sounds seductively like Darwinian natural selection. The seduction should be resisted, or should at least ring alarm bells. Superficial resemblances can be actively misleading.

The fact of our own existence is almost too surprising to bear. So is the fact that we are surrounded by a rich ecosystem of animals that more or less closely resemble us, by plants that resemble us a little less and on which we ultimately depend for our nourishment, and by bacteria that resemble our remoter ancestors and to which we shall all return in decay when our time is past. Darwin was way ahead of his time in understanding the magnitude of the problem of our existence, as well as in tumbling to its solution. He was ahead of his time, too, in appreciating the mutual dependencies of animals and plants and all other creatures, in relationships whose intricacy staggers the imagination. How is it that we find ourselves not merely existing but surrounded by such complexity, such elegance, such endless forms most beautiful and most wonderful?

The answer is this. It could not have been otherwise, given that we are capable of noticing our existence at all, and of asking questions about it. It is no accident, as cosmologists point out to us, that we see stars in our sky. There may be universes without stars in them, universes whose physical laws and constants leave the primordial hydrogen evenly spread and not concentrated into stars. But nobody is observing those universes, because entities capable of observing anything cannot evolve without stars. Not only does life need at least one star to provide energy. Stars are also the furnaces in which the majority of the chemical elements are forged, and you can't have life without a rich chemistry. We could go through the laws of physics, one by one, and say the same thing of all of them: it is no accident that we see . . .

The same is true of biology. It is no accident that we see green almost wherever we look. It is no accident that we find ourselves perched on one tiny twig in the midst of a blossoming and flourishing tree of life; no accident that we are surrounded by millions of other species, eating, growing, rotting, swimming, walking, flying, burrowing, stalking, chasing, fleeing, outpacing, outwitting. Without green plants to outnumber us at least ten to one there would be no energy to power us. Without the ever-escalating arms races between predators and prey, parasites and hosts, without Darwin's 'war of nature', without his 'famine and death' there would be no nervous systems capable of seeing anything at all, let alone of appreciating and understanding it. We are surrounded by endless forms, most beautiful and most wonderful, and it is no accident, but the direct consequence of evolution by non-random natural selection - the only game in town, the greatest show on Earth.

Was this article helpful?

0 0


  • petros
    What does evolvability mean?
    9 years ago

Post a comment