摘录

Chapter One The imprint of evolution Introduction We live on a wonderful planet that not only teems with life but shows a marvellous exuberance of form and variety. In comparison with the size of the Earth its living skin (the so-called biosphere) may be thin, but it is by no means negligible. From high in the atmosphere, where ballooning spiders wafted aloft on their silk-strings have been trapped at heights of more than 4500 m and birds such as condors cross tropical storms at altitudes well above 6000 in, via the oceans and the green continents, to deep within the Earth's crust where bacteria are known to live at depths of at least several kilometres, life is pervasive.     Nobody knows the precise total of species that presently inhabit the Earth, nor how many once existed but are now extinct. There could quite easily be twenty million species alive today, and the number of extinct species must run into the hundreds of millions, if not the billions. Within this vast plenitude it is perhaps rather surprising that there is only one, unique species that can understand a single word of this book. This species, which is of course ourselves, is uniquely privileged: not only can we understand something of our origins, but we are the first animals ever to have looked at the stars and seen anything more than distant pin-pricks of light.     Because, in some ways, we are utterly different from any other form of life that has ever evolved, how do we know that our origins and history are to be traced here on Earth rather than as extraterrestrial immigrants? The reason is simple: our evolutionary pedigree is stamped on every feature and permeates the entire fabric of our bodies. Some aspects of our history are of comparative recency. For example, our ability to walk upright (the bipedal stance) was achieved only about four and a half million years (Ma) ago. The astonishing increase in our brain size, even in comparison with the closely related apes, is yet more recent. The basic structure of our arms and our legs, including the characteristic five fingers and toes (technically the pentadactyl limb), can be traced back over hundreds of millions of years. Indeed, it is now possible to study fossils, including some collected from Devonian rocks (about 370 Ma old) in east Greenland, that indicate how the fins of aquatic fish were transformed into the limbs of the first terrestrial vertebrates. Similarly, although our brains are unique in their mental and spiritual faculties, the basic structure of the brain is easily identifiable in primitive fish. This arrangement must have evolved at least 500 Ma ago. But our evolutionary history is much more deeply encoded than organs such as limbs and brains. In many ways our basic biochemistry is little different from that of the bacteria. These steps in evolution were achieved thousands of millions of years ago. Not only do we and bacteria both use DNA for replication, but special proteins (the histones) that surround the strand of DNA and assist with keeping it stable and in the correct configuration are very similar in their sequence of building blocks (the amino acids) in all life. This is simply because they play a fundamental role in maintaining the proper function of the DNA; most alterations are automatically fatal.     It is, however, self-evident, even if the histone proteins are almost invariant in their structure, that life itself has not remained at the level of bacteria. The world is full, not only of bacteria, but also of animals as different as cranes, whales, oysters, and sharks, not to mention the plants, fungi, and single-celled organisms such as Amoeba. This book is not directly concerned with the origins of any of these creatures, or indeed ourselves. Rather it is an exploration of how a single unit of rock, from the west of Canada and known as the Burgess Shale, has placed the history of life, and so by implication Man's place in the scheme of evolution, in a new set of contexts.     What then is this Burgess Shale and why is it regarded as so important? How it was discovered, who worked on it, what scientific mistakes they inevitably made, how much remains to be learnt, and whether the whole concept of evolution in the Darwinian framework now needs to be radically reconsidered will all be considered in the rest of this book. The Burgess Shale is a thin unit of rock. The outcrop itself, in a small quarry on the side of a hill, is rather drab and unremarkable, but any palaeontologist would want to work there for two reasons. One is seemingly trivial: even if the quarry looks very ordinary, the Burgess Shale occurs in some of the most beautiful scenery in the world, in the Main Ranges of the Canadian Rocky Mountains. Looking from the quarry, as far as the eye can see, there are snow-capped mountains, glaciers, turquoise-coloured lakes, and forests set in wilderness. If one has to collect fossils, one might as well collect them here! The second reason is that the Burgess Shale is no ordinary fossil deposit. Here, by as yet largely unknown mechanisms, the processes of rotting and decay have been largely held in abeyance so that the true richness of ancient life is revealed: not only are there animals such as trilobites and molluscs with tough, durable skeletons, but completely soft-bodied animals are also preserved. These remarkable fossils reveal not only their outlines but sometimes even internal organs such as the intestine or muscles.     The Burgess Shale is not unique, but for those who study evolution and fossils it has become something of an icon. It provides a reference point and a benchmark, a point of common discussion and an issue of universal scientific interest. Just as Darwin's finches from the Galapagos Island exemplify the recognition of the central role of adaptation, or the laboratory fly Drosophila stands as a symbol for the profound successes of molecular biology, so the Burgess Shale is becoming the icon for those who study the history of life. But before we begin to understand what the riches of the Burgess Shale mean, both to evolution and the scientific method, it is essential to place it in a wider context. By obtaining a sense of its place in the unfolding drama of life, set in an ecological theatre, so we can understand why it has become one of the leading players. Evolution: why no consensus? All science is embedded in a framework, which provides the points of reference and a necessary stability to our enterprise. Not surprisingly, many aspects of the framework remain little changed for decades, and on a day-to-day basis are accepted and usually remain unchallenged. For biology it has been famously observed that nothing makes sense unless considered in the context of evolution. The fact of organic evolution in itself is not in dispute. This is because in essence the Darwinian formulation of descent through time and co-occurring modification of the organisms, usually registered in the fossil record by anatomical changes, seems to be unanswerably correct. Once there were only bacteria; now they share the planet with millions of other types of life. Separate and special creation of each and every species is a logical alternative, and in itself need not be beyond reason. Nevertheless, the study of comparative anatomy, behaviour, molecular biology, and the fossil record give no support to any such model of recurrent creation.     So if we accept a tree of life, arising from a single ancestor approximately four thousand million years ago, why does the apparently simple fact of organic evolution excite continuing debate and disagreement? What is it that is in dispute? At heart there are two areas of contention: those of mechanism and those of implication. The first is a scientific problem, the second is metaphysical. Our immediate concern here is with the aspects of evolutionary theory (as presently portrayed) that are relevant to the Burgess Shale. As with most areas of science, the argument proceeds by reference to examples. The story of the Burgess Shale therefore epitomizes many aspects of the debate on evolution, but this extraordinary fauna is nevertheless no more than a convenient vehicle that embodies the wider principles that are at stake.     A recurrent difficulty in discussions on organic evolution is that schools of thought are too often polarized, although this is understandable because of the need to solve tractable problems that need to be stated in high circumscribed language. Nevertheless, in all the debates and disagreement, it seems rather extraordinary that for the most part it is almost forgotten that evolution is a historical process. In part it is accessible from the fossil record, and the analogies with the study of human history are clear. For example, if I wish to know more about the history of a college in Cambridge I can spend a rewarding time in the archives, aware that not all documents are decipherable and some may have been lost by fire, flood, or worm. Much can also be learnt, however, from simply studying the present order, be it of the buildings or the nature of its society. Here, too, there will be a clear historical stamp. It is all the more remarkable that the pitfalls and fallacies that the well-trained historian teaches others to avoid do not seem to be utilized by those investigating the parallels in the history of life. In a society stricken by post-Saussurean relativism it is also too often forgotten that history had a unique course, and that in principle it is knowable.     What then of evolutionary mechanism? In brief, there seem to be three main problems to consider. It is widely, although not universally, agreed that central to the evolutionary process is the splitting of lineages, with at least one of the descendant forms differing materially from the ancestral type. Most biologists identify this process as one of speciation, the formation of new species. In classical biology this aspect of evolution has been construed in terms of mechanisms that promote the isolation of groups of individuals (populations) and thereby, at some subsequent stage, an inability to interbreed or at least produce fertile offspring. The frequency of hybridization, as well as the possibility for the transfer of genetic material between species, perhaps by the agency of bacteria, demonstrate that species need not be watertight entities, at least genetically. Forms transitional between species can be observed today, and can be inferred to have existed in the past. Nevertheless, the net result is very far from a seamless tapestry of form that would allow an investigator to read the Tree of Life simply by finding the intermediates--living and extinct--that in principle connect all species. On the contrary, biologists are much more impressed by the discreteness of organic form, and the general absence of intermediates.     Here, therefore, lies an important area of tension in the study of evolution. On the one hand the diversity of life can be read from an essentialist point of view, one whose vocabulary will include words such as body plan (or Bauplan) and archetype. In their more far-flung moments of comparison, proponents will take an effectively Platonic view that organic form reflects some sort of universal order, akin to the ideal solids of Platonic metaphysics. In this essentialist view the implication is that organic diversity is imposed, rather than evolved. It will also be clear that the essentialist views could be compatible with those that seek evidence for special creation in organic form. In marked contrast is an alternative viewpoint of evolutionary processes that might be linked to the famous Heraclitean flux of continuous change. In one sense this must be uncontroversial because, barring appeals to hopeful monsters testing their saltatory abilities, the facts of evolution point to building up on previously available organic designs in a gradualistic manner. Whatever is in dispute about evolution, it is not the derivation of one type from another. But if we accept the reality of transitions do we not have to explain why large sections of potential morphospace remain unoccupied? If this is the correct analysis, as indeed it appears to be, then a more profound problem emerges as to whether such vacancies reflect lack of chance or opportunity, or whether (as seems more plausible to most Darwinian biologists) some zones of organic form (or morphospace) are effectively impossible to colonize because any organism occupying them would be seriously maladapted. In an ideal case such regions of morphospace are described mathematically. One such example is given in Chapter 8, where an example is taken from the extinct trilobites. In many other cases a precise mathematical description that defines the morphospace occupied by a group of organisms remains a very challenging prospect, but headway has been made. One of the best-known examples concerns the growth and hence geometry of the shells secreted by the molluscs, a group familiar from animals such as the garden snail and edible mussel. Although not immediately apparent, the geometry of nearly all mollusc shells can be reduced to several simple equations that together describe their various shapes. What this means is that any point in mollusc shell morphospace can be defined according to a given solution of these equations. Not surprisingly, when applied to the real world such an analysis shows some regions of morphospace to be thickly populated by shell types that are relatively familiar. Other zones, however, are more or less empty. In these latter cases, the equations can be readily used to visualize the hypothetical shell shape we would find if this region of morphospace was occupied, but somehow they look `wrong'. Such regions of morphospace housing these aberrant shells need not be entirely vacant, but the most likely explanation is that such forms are (or were) at a serious selective disadvantage for reasons such as mechanical weakness or vulnerability to predation.     To return to the specifics of organic evolution. It is generally accepted that the origins of divergence of form are coincident with the processes of speciation itself. Although it may be a mistake to think of speciation as a single process, the end results of these processes seem to be much the same. Let us accept then, if only for the sake of the argument, that not only are species discrete entities, but that they arose from pre-existing forms from which they differed in some material aspect. The central question is: are the processes of speciation in themselves sufficient to explain the pattern of life that we see today or at any time in the geological past? For nearly all biologists the fact of speciation is not in dispute, but its role in driving evolution is much more contentious. For those who do not accept speciation as the main motor of organic diversity, there are broadly two approaches. There are those who look to the molecular dynamics of the genome, as against those who seek some wider view that transcends the species. Thus, according to a number of molecular biologists the crux of the investigation needs to move to the genome and the reorganization and reshuffling of pieces of DNA, the molecular units of heredity. It is certainly realized that the genome is much more dynamic than was once thought. For example, there are large variations in the amount of DNA in different species, and it is still far from clear why some organisms have such huge excesses in DNA. There is little connection to complexity: humans for example have relatively modest amounts of DNA in each cell, but to dismiss--as some have done--the apparently excess DNA as `junk' may be too simplistic. Not only can the amount of DNA in the chromosomes be dramatically increased, but in addition genes can be shuffled, moved around, or duplicated. There is also evidence for transfer of genetic material within the cell, notably moving DNA from the organelles known as mitochondria (which house their own separate circular chromosome) to the main storehouse of DNA in the chromosomes housed in the nucleus. In itself this activity need not be under the scrutiny of natural selection, even if the end result, the expressed phenotype, is moulded and constrained by the classical Darwinian principles of variation followed by selective culling. At the other end of the spectrum are those who argue that it is the evolutionary processes operating above the level of species that are unjustly neglected. There has been particular interest in a mechanism referred to as species selection, which in outline states that a propensity to speciation, in itself unrelated to the operation of natural selection, will favour one clade, that is, a set of species sharing a common ancestor, over another clade. While the principle of species selection appears to be logical, there are to date very few case-examples to suggest that it is of particular significance.     There are other aspects to organic evolution that are certainly not ignored, but perhaps still receive insufficient emphasis. One is the influence of the environment. This would seem to be unremarkable, until it is realized that much of current thinking seems to be firmly embedded in a uniformitarian framework, that is, it assumes that present-day conditions are a sufficient guide to understanding past worlds. In some ways this must be true: the sun shines, water is wet, and things fall out of trees. But in other ways the Earth has clearly changed dramatically. It appears that in the past 600 million years the composition of the atmosphere, notably in terms of oxygen and carbon dioxide, has changed significantly. Times of elevated oxygen levels, for example, coincide with gigantism and the development of flight in some animals. There is a suspicion that there is a causal connection. Here is another example. Further back in time the Moon was probably much closer to the Earth. Because of the inverse square law of gravitational attraction, the proximity of the Moon would then have generated immense tides. What effect did these have on primitive life? Could this explain, in part, the sluggishness of organic evolution at this time? This might not be the only environmental constraint. Some workers have suggested that early in the history of the Earth surface temperatures were significantly elevated, and this too could have exerted a powerful brake on organic diversification.     Nor need the controls on evolution be exerted by environment alone. Barring sudden catastrophes, such as the arrival of a giant meteorite, most environmental factors will change at an imperceptible rate when compared with the generation times of living populations. But evolution proceeds not only in a real physical world, but in a biological arena. It would be simplistic to imagine that species are `locked in' to an ecological framework, but the communities and biomes which they occupy must exert some degree of constraint.     It will be clear by now that although the Darwinian framework provides the logical underpinning to explain organic evolution, the actual pattern of life we observe may require a more complex set of explanations. Those who believe that their viewpoint is being neglected may be strident in their claims. Perhaps one reason for the continuing debate is that as a whole the various mechanisms proposed are each eminently reasonable. It is the problem of deciding if one such mechanism deserves primacy of effect, or whether the question `Why do organisms evolve?' is unanswerable until one specifies the mechanism and the level at which it may operate. In such a large and complex field, the main strands of debate, and sometimes enquiry, are accordingly difficult to disentangle, not least because among some of the main proponents there are often broad areas of agreement. Indeed, some generate an aura of apparent accommodation by stressing their plurality of approach. On closer examination, however, this sometimes transpires to be skin-deep. Moreover, those with ideological training know that the tactics of persuasion may be assisted by the invention of key phrases that demonize the opposition.     Who then are the main proponents? Because of his earlier discussion of the Burgess Shale fauna, it is essential to review the contributions of Stephen Gould. But before doing so it is necessary to introduce briefly those who would regard their view of the evolutionary process as more or less antithetical to that of Gould. This latter group can be labelled, I think fairly, as hard Darwinists. One spokesman, Daniel Dennett, has elevated the Darwinian method to what is effectively a universal principle. The acid test of such a claim is whether such a formula can explain what are presently regarded as the most fundamental and least tractable of problems, notably those of cosmology and the early history of the Universe and the onset of consciousness. For many this is taking the principle too far, and it is certainly the case that the entire philosophy is strongly materialist. In terms of organic evolution nowhere is this more evident than in the vigorous advocacy of Richard Dawkins. In a series of polemical, but carefully argued and vividly expressed books, Dawkins has unremittingly pursued the consequences--as he sees them--of the Darwinian world picture. Although set in an adaptationist landscape, a rolling and sometimes mountainous terrane that encompasses not only form and function but also behaviour, his fundamental point of reference is the primacy of the gene. In this way, Dawkins takes a highly reductionist approach. Not surprisingly, however clear the articulation, this programme has generated controversy and unease because of a sense in which the richness and diversity of evolution are being forced into an atomistic mould. Dawkins would probably reply that he is only seeking the underpinnings of the evolutionary process, upon which all else depends.     It is certainly the case that recent research into the developmental processes in animals has been little short of spectacular. At first sight these results seem to be consistent with the primacy of the gene. In a number of instances it is clear that a specific gene is associated with the expression of a complex anatomical feature. One of the best-known examples involves a so-called master-control gene which plays a key role in the formation of eyes. In a classic but disturbing experiment, the application of this gene to the fruit-fly led to an ectopic expression, that is, to eyes growing on various parts of the body. But this and similar genes hold further surprises that suggest the story to be more complicated. First, it transpires that the same gene (Pax-6 and its homologue eyeless) is employed not only in flies and other insects to build their characteristic compound eyes, but also in vertebrates. The eyes with which you read this page result in one sense from the activity of the same gene. Yet, despite the fact that both are light-receiving organs, there are profound differences between the eyes of fly and Man. On further reflection this need not surprise us. Most probably the Pax-6 gene is very ancient. It almost certainly predates the animal, presumably some sort of worm, that about 600 million years ago represented the common ancestor of flies and humans. Indeed Pax-6 may predate the earliest animals. This is because its function is to construct a light-sensitive unit, and such structures are well known in a number of the more primitive single-celled organisms whose origins almost certainly predate the animals. Equally important the recognition of Pax-6 in arthropods (flies) and vertebrates (humans) is good evidence that they are indeed related, but it tells us nothing about the manifest differences between the eyes with which we see the fly, and the eyes of the fly which observe us as we advance with rolled newspaper in hand.     It is in this manner that Dawkins's world view is not so much wrong, as simply seriously incomplete. While few doubt that the development of form is underwritten by the genes, at the moment we have almost no idea how form actually emerges from the genetic code. In his enjoyable book The shape of life the American evolutionary biologist Rudy Raff is bald in his assessment: `The central problem is finding the mechanisms that connect genes and developmental processes to morphological evolution' (p. 430). One puzzling aspect, for example, is that species with very similar adult forms may reach this final stage via markedly different developmental pathways. These so-called trajectories may in themselves have adaptive significance, and no doubt different sets of genes swing into action at different times. In addition, seemingly major contrasts in anatomical arrangement may well depend on trivial genetic differences. Until, however, we learn what these are, we shall remain uninformed about the actual mechanisms whereby the shape of life is moulded. It is certainly difficult to see how the severely reductionist approach of Dawkins will continue to provide the most satisfactory strategy. Indeed, what has quite unexpectedly emerged is how seemingly very different organisms have in common fundamentally the same genetic information. Here is perhaps the central paradox of genes and evolution: vast contrasts in morphology and behaviour need have no corresponding differences in the genetic code.     Perhaps a suitable analogy to explain the short-falls of Dawkins's account of evolution is to think of an oil painting. In this analogy Dawkins has explained the nature and range of pigments; how the extraordinary azure colour was obtained, what effect cobalt has, and so on. But the description is quite unable to account for the picture itself. This view of evolution is incomplete and therefore fails in its side-stepping of how information (the genetic code) gives rise to phenotype, and by what mechanisms. Organisms are more than the sum of their parts, and we may also note in passing that the world depicted by Dawkins has lost all sense of transcendence.     In such a multifarious subject as evolution, it is certainly possible to identify camps (and outposts), but it is less easy to arrange them into a linear spectrum, let alone a simple polarity. Yet, if there is some sort of antithesis to Dawkins's portrayal of evolution, it is a yet stranger world inhabited by Stephen J. Gould, who rivals Dawkins as a popularizer of evolutionary biology. At first sight Gould's construction is much richer, especially in its appeals to a plurality of mechanisms and forces. But it is also a less constant world, or at least one where emphases and priorities shift. The world picture offered by Dawkins, as I have suggested above, is not so much wrong as simply too narrow and one-dimensional. The one presented by Gould is much more difficult to encompass, but despite its apparent vitality, I would argue that it is much more deeply flawed. Because the faunas of the Cambrian, and especially the Burgess Shale, have taken a key role in some of Gould's more recent perorations, notably in the book Wonderful life, it is necessary to take into account the general background of his view of life, and so its evolution.     To start with, Gould does not attempt to deny the importance of the Darwinian explanation. And indeed why should he? Some of the most cogent and readable explanations of these evolutionary principles are compelling and fascinating, especially those concerning the manner in which complex structures are `jury-rigged' from pre-existing structures in an apparently contrived way, the nature of which clearly reveals the deep historical imprint of evolutionary activity. But Gould has also not ceased to champion the notion that the Darwinian explanation is in some way incomplete. It is hardly surprising that he has found himself at loggerheads with Dawkins. Again and again Gould has been seen to charge into battle, sometimes hardly visible in the struggling mass. Strangely immune to seemingly lethal lunges he finally re-emerges. Eventually the dust and confusion die down. Gould announces to the awestruck onlookers that our present understanding of evolutionary processes is dangerously deficient and the theory is perhaps in its death throes. We look beyond the exponent of doom, and there standing in the sunlight is the edifice of evolutionary theory, little changed. One source of unease in Gould's writing is what appears to some people as the fine line between argument and rhetoric. Thus, a favourite rallying cry of his was to label the neo-Darwinian programme, largely built on the population genetics of Morgan and Dobhansky and the mathematics of Fisher, as hopelessly sclerotic: what Gould famously labelled as `the hardening of the synthesis'. This was a master stroke of invective, and is perhaps reminiscent of the political tactic of picking a resonant phrase to box in and demonize one's opponents. But is it a fair comment? There was only a `hardening' inasmuch as what the neo-Darwinian school set out to do was immensely successful, and was pursued with vigour. Did it stifle research? If neo-Darwinians turned their collective back on a much-vaunted plurality of alternative evolutionary mechanisms, were they ultimately so unwise? It is significant that the recent dramatic advances in developmental biology can be directly traced to the painstaking work of these earlier neo-Darwinian geneticists. Not only that, but the repeated invitations to reinstate such individuals as Richard Goldschmidt and Otto Schindewolf from being isolated voices in the wilderness to occupy a favoured place in the pantheon of evolutionary biology have quite simply failed. That both these individuals made important contributions is not in dispute, but at the time were they ever a serious threat to our understanding of the evolutionary process? Although less commented upon than Goldschmidt, whose work on butterflies has been overshadowed by his celebrated leap into macroevolutionary thought by the agency of his much-discussed `hopeful monsters', Schindewolf is also an interesting case history. Embedded in Spenglerian cyclicity, whereby groups of organisms contained the seeds of their disaster and from high triumph descended into decadence and rottenness, his scientific influence in Germany was enormous, and baneful. A rather sinister combination of autodictat and adherence to a flawed philosophy led German palaeontology into a cul-de-sac of sterile macroevolutionary speculation and an anti-Darwinian attitude that persisted for many years after the overthrow of the Nazis.     Such is the complexity of evolutionary discussion that it would not be fair to dismiss what are now generally thought of as hopeless cases without a fair hearing. That evolution is rich in unsolved problems is not in dispute. It is certainly true that Gould's enthusiastic promulgation of various alternatives to evolutionary orthodoxies has made the guardians of neo-Darwinism look more carefully at their received truths. These alternatives have generated healthy debate. There needs, however, to come a time not only for summation and the taking of stock, but also to enquire whether other problems of evolution remain neglected. Take the case of adaptation, a key element in the Darwinian framework. That it exists is not in dispute, but is it crucial to our wider understanding of evolution? After all, if combinations of characters and traits can `slip past' the scrutiny of natural selection, then perhaps the architecture will reveal unexpected riches. And it was by a characteristically inventive, but as we can now see flawed, metaphor that Gould started a debate on the importance of adaptation that now looks to be increasingly misplaced. He fired the first shot in his paper (with R.C. Lewontin) on the spandrels of the Doges' chapel in Venice, the famous San Marco. (A spandrel can be defined in more than one way, but here it can be regarded as the roughly triangular space between the shoulders of two adjacent arches and the horizontal line immediately above their heads.) The argument that Gould and Lewontin put forward was that just as these architectural features are incidental to the design of the building, so organisms also house their own `spandrels', which are similarly without adaptational significance. A supposed architectural by-product was taken as the introduction to a polemic on the dangers of viewing the world through adaptationist spectacles. But in fact Gould and Lewontin's analysis is fundamentally flawed. Spandrels are very far from being incidental by-products of construction and are central to design and safety. It may be no accident that the almost universal human admiration of adaptation in the organic world is in some ways echoed in such buildings as San Marco. The spandrels, or more properly pendentives, house some of the glowing and mysterious Byzantine-inspired mosaics that draw the observer towards a deeper contemplation of Christian faith. Moreover, is not much of our disenchantment with the barbarity of much recent architecture due to this banishment of the numinous?     The case of the spandrels is one of the better known of Gould's evolutionary perspectives, and is perhaps overshadowed only by the hypothesis of punctuated equilibria. Nevertheless, despite some shifts in emphasis, the underlying ideological agenda of Gould has always been fairly clear. Even where there has been a shift in thinking, it might be argued that in general the discussions were reflecting a particular world-view that at the least was sympathetic to the greatest of twentieth-century pseudo-religions, Marxism. Thus at one stage an influential group of American biologists was interested in exploring a so-called nomothetic view of evolution. This was an attempt, perhaps futile, to seek general laws of evolution, which if discovered might allow the practitioners to claim that evolutionary biology was a `hard' science, comparable in some sense to chemistry and physics. As is well known, the Marxist agenda has long sought `laws' of history, principally linked to certain inevitable outcomes that strangely favoured those fortunate enough to have formulated the `laws' in the first place. There is, of course, no suggestion that the hegemony of an ideology is to be transferred to the inevitability of a certain view of biology. My point rather is that the nomothetic investigation of historical sciences may reveal some interesting parallels. In any event, so far as evolutionary biology is concerned this programme has been effectively abandoned; apart, that is, from a small group of anti-Darwinians who have pursued the enterprise in the rather different direction of explaining organismal form by various underlying `forces'. In the meantime, of course, there has been a spectacular growth of interest in the operation of mathematical systems of non-linear dynamics, popularly referred to as chaos theory.     In more recent years Gould has promulgated a rather different set of notions that emphasize the role of the contingent in evolution. At first sight it is quite difficult to decide whether any of this needs to be taken seriously--until, that is, the underlying message is decoded. It is indeed somewhat surprising that the operation of contingency needs any comment at all. After all, if St Thomas Aquinas had no difficulty in reconciling the order of a Universe stemming from the Act of a Creator, part of which entailed a contingent world, then we might wonder how those involved with the more mundane role of explaining evolution could sense that contingent events had been an overlooked part of the puzzle. If thorough-going theists, who traditionally have been supposed to be hostile to the scientific theories of evolution, are content to accept contingency, then one might presume that its operation in the history of life would pass unremarked. And so it might, until it was seized upon by Gould as a point worth serious discussion. In brief his argument, largely using the Burgess Shale faunas, was that the range of variation in the Cambrian was so huge and the end results in terms of the diversity of today's world so restricted that the history could be regarded as one colossal lottery. Forget the big battalions, inspired leadership, the idiosyncrasy of genius, the professionalism of the academies, or any of the other factors that are routinely used to explain the twists and turns of human history, and by analogy the history of life: quite possibly they are relevant to the human condition, but no such correspondence existed in the natural world. Here chance reigned supreme, with the corollary that what to us constitutes the utterly familiar was in principle no more inevitable than a million other outcomes, ones in which humans would assuredly play no part. So much for the flights of rhetoric. Here, nevertheless, we are with one state of affairs--the world around us. How could we ever show that a plenitude of alternatives was equally likely, with the important corollary that nothing like us humans would be there in this imaginary world, either to ponder or to celebrate?     I presume that the best test of this supposition would be the discovery of a distant planet, sufficiently Earth-like to support some sort of animal life. In Chapter 9 I note that the likelihood of extraterrestrial life itself, let alone anything remotely like a human, may be much more remote than is popularly supposed. But in the immediate terms of discussing the outcomes of alternative histories, quite possibly with only marginal differences in the starting conditions, the question of whether there is or is not extraterrestrial life may not be too material to the argument. In one sense the experiment of alien life has been carried out, but here on Earth. Thus, although there is of course an evolutionary continuity in the history of life, it is also the case that not only are nearly all the species that have ever lived now extinct, but entire ecosystems have also vanished. In these past worlds there was much that was novel and has no counterpart today. But it is also true that much is familiar. This is not so much to do with evolutionary continuity, but the phenomenon known as convergence. Any textbook of evolution that fails to mention convergence would be guilty of serious dereliction. Yet despite the classic examples, which vary from the anecdotal to the closely argued, the study of convergence and the constraints of form, have, I believe, never been the subject of a single synthesis. There are several reasons for this. One is its simple ubiquity: convergence is taken for granted. Another is the problem of formulating a precise metric of convergence. Famously, the marine reptiles known as ichthyosaurs are remarkably similar to the living dolphins; but are the convergences only superficial or of deep significance? Convergence is seldom precise. In addition, to identify convergence one must know the evolutionary tree that depicts both the interrelationships and ancestral conditions. But this can only be done on the basis of similarities of organization, be they anatomical, behavioural, or molecular. Thereby lies the risk of becoming trapped in a circular argument: are organisms similar because they have converged or because they are descendants of a common ancestor? In terms of specifics this remains a very serious problem, but in terms of generalities the problem evaporates because no matter what evolutionary tree is chosen, convergent features almost invariably emerge. The reason for discussing convergence here is that its recognition effectively undermines the main plank of Gould's argument on the role of contingent processes in shaping the tree of life and thereby determining the outcome at any one time. Put simply, contingency is inevitable, but unremarkable. It need not provoke discussion, because it matters not. There are not an unlimited number of ways of doing something. For all its exuberance, the forms of life are restricted and channelled.     For the great majority of biologists such a conclusion will hardly be surprising. The agenda, however, once again is ideological, because the discussions on contingency versus constraint seem to be more to provide the background and focus of a very specific problem, that is the rise of human intelligence. Gould's view is unequivocal. The likelihood of Man evolving on any other planet is extraordinarily unlikely. To paraphrase: if the history of evolution were to be repeated, the world would teem with myriad forms of life (note that the contingent likelihood of the origin of life itself goes through on the nod), but certainly no humans. As stated, this seems to be entirely unremarkable, although again it presupposes that the constraints are weak. It is not, however, the point. What we are interested in is not the origin, destiny, or fate of a particular lineage, but the likelihood of the emergence of a particular property, say consciousness. Here the reality of convergence suggests that the tape of life, to use Gould's metaphor, can be run as many times as we like and in principle intelligence will surely emerge. On our planet we see it in molluscs (octopus) and mammals (Man). It might still be objected that the properties expressed in Man have a uniqueness without precise parallel. This may be a distortion of the time in which we fortuitously find ourselves; what was rare in the last four thousand million years of evolutionary history might be common in the next four thousand million years. Weak support for this argument might come from the most closely related species to us, the Neandertals. Perhaps independently they developed some sort of sense of an afterlife, at least to judge from their practice of deliberate burial. Materialists will scoff at this as a shared delusion, but there are metaphysical alternatives that are perhaps more fruitful.     But Gould's arguments on the quirkiness of human intelligence are not only presented as part of an evolutionary argument, but also I believe to buttress an ideological viewpoint. In brief, his assessment of Man as an evolutionary accident is to lead us into a libertarian attitude whereby, by virtue of a cosmic accident, we, and we alone, have no choice but to take responsibility for our own destiny and mould it to our desire. At the very least, the activities of the last century as one of unrestricted political experimentation should give us pause for thought. The implication of an evolutionary process transcending the scientific evidence does indeed provide a metaphysic, albeit one that is etiolated and impoverished, but it should be decisively rejected. We do indeed have a choice, and we can exercise our free will. We might be a product of the biosphere, but it is one with which we are charged to exercise stewardship. We might do better to accept our intelligence as a gift, and it may be a mistake to imagine that we shall not be called to account.     As I noted above, we muddy the waters of the debate if we fail to acknowledge that the processes of evolution have metaphysical implications for us. This is because uniquely there is inherent in our human situation the possibility of transcendence. The fact that we arrived here via an immensely long string of species that originated in something like Pikaia rather than some other crepuscular blob is a wonderful scientific story, but it is hardly material to our present condition.