* * *
{i}
Copyright © 1996 by Richard Dawkins
Original drawings copyright © 1996 by Lalla Ward
All rights reserved
First American Edition 1996
Printed in the United States of America
The text of this book is composed in Centaur. Composition by Justine Burkat Truhey
using Adohe Pagemeaker 6.0. Manufacturing by The Haddon Craftsmen, Inc.
Book design by Margaret M. Wagner
Library of Congress Cataloging-in-Publicaiion Data
Dawkins, Richard, 1941—
Climbing mount improbable / Richard Dawkins; original drawings by Lalla Ward.
p. cm.
Includes bibliographical references and index.
ISBN 0-393-03930-7
1. Natural selection. 2. Evolutionary genetics. 3. Morphogenesis. I. Title.
QH375.D376 1996
575.01'62 — dc20
96-19138
CIP
W. W. Norton & Company, Inc. 500 Fifth Avenue, New York, N.Y. 10110
http://www.wwnorton.com
WW Norton & Company Ltd., 10 Coptic Street, London WCIA IPU
1 2 3 4 5 6 7 8 9 0
* * *
{iv}
COLLEGE LIBRARY
For Robert Winston,
a good doctor and a good man
* * *
{v}
Acknowledgements ix
Picture Acknowledgements xi
1 Facing Mount Rushmore 3
2 Silken Fetters 38
3 The Message from the Mountain 73
4 Getting off the Ground 108
5 The Forty-fold Path to Enlightenment 138
6 The Museum of All Shells 198
7 Kaleidoscopic Embryos 224
8 Pollen Grains and Magic Bullets 256
9 The Robot Repeater 276
10 A Garden Inclosed 299
Bibliography 327
Index 333
* * *
{vii}
Acknowledgements
THIS BOOK GREW OUT OF MY ROYAL INSTITUTION Christmas Lectures, televised by the BBC under the general title Growing Up in the Universe. I have had to abandon that title because at least three other books have since appeared with almost identical names. Moreover, my book itself has grown up and changed, so it is no longer fair to call it the book of the Christmas Lectures. Nevertheless I should like to thank the Director of the Royal Institution for honouring me with the invitation to join the historic lineage of Christmas Lecturers going all the way back to Michael Faraday. Bryson Gore of the Royal Institution, together with William Woollard and Richard Melman of Inca Television, influenced the lectures greatly, and traces of their influence will still be found in this greatly transformed and enlarged book.
Michael Rodgers read and constructively criticized early drafts of more chapters than are here printed, and advised decisively on the reconstruction of the whole book. Fritz Vollrath and Peter Fuchs gave expert readings of Chapter 2, while Michael Land and Dan Nilsson did the same for Chapter 5. All four of these experts gave generously of their knowledge when I tapped it. Mark Ridley, Matt Ridley, Charles Simonyi and Lalla Ward Dawkins read the whole book in a late draft and provided helpful criticism and reassuring encouragement in needful proportions. Mary Cunnane of W. W. Norton and Ravi {ix} Mirchandani of Viking Penguin showed generous tolerance and big-hearted judgement as the book grew, took on a life of its own and finally shrank again to more manageable scope. John Brockman lurked encouragingly in the background, never interfering but always ready with support. Computer experts are heroes, too often unsung. In this book I have used the programs of Peter Fuchs, Thiemo Krink and Sam Zschokke. Ted Kaehler collaborated with me in conceiving and writing the difficult Arthromorphs program. In my own suite of ‘watchmaker’ programs I have frequently benefited from the advice and help of Alan Grafen and Alun ap Rhisiart. The staff of the Zoological and Entomological Collections of the University Museum at Oxford lent specimens and expert advice. Josine Meijer was a willing and resourceful picture researcher. My wife, Lalla Ward Dawkins, did the drawings (though not the layouts) and her love of Darwinian Creation shines through every one of them.
I should like to thank Charles Simonyi, not only for his immense generosity in endowing the post in Public Understanding of Science which I now hold at Oxford, but also for articulating his vision — which coincides with mine — of the craft of explaining science to a large audience. Do not talk down. Try to inspire everybody with the poetry of science and make your explanations as easy as honesty allows, but at the same time do not neglect the difficult. Put extra effort into explaining to those readers prepared to put matching effort into understanding. {x}
* * *
>>
Picture
Acknowledgements
Drawings by Lalla Ward: 1.7, 1.9, 1.10, 1.13, 1.14, 2.9, 3.1, 3.3, 4.2, 4.3, 4.4, 4.5, 4.6, 4.7, 5.1, 5.15, 6.3, 6.4, 6.10, 6.13, 6.15, 7.3, 7.8, 7.15a, 7.16, 8.2, 8.3, 8.6; 1.2 (after Holldobler and Wilson); 1.3 (after Wilson); 1.11 (after Eberhard); 2.6 (after Bristowe); 5.30 (after M. F. Land); 7.10 (after Brusca and Brusca); 7.11 (after Collins Guide to Insects); 7.17 (after Brusca and Brusca); 10.6 (after Heijn from Ulenberg).
Computer-generated images by the author: 1.14, 1.15, 1.16, 5.3*, 5.5*, 5.6*, 5.7*, 5.9*, 5.10*, 5.11*, 5.12, 5.20*, 5.28, 6.2*, 6.3*, 6.5, 6.6, 6.8, 6.11, 6.12, 6.14, 7.1, 7.9, 7.12, 7.13, 7.14 (images marked with an asterisk redrawn by Nigel Andrews); by Jeremy Hopes 5.13.
Heather Angel: 1.5, 1.11b, 5.21, 8.1. Ardea: 1.8 (Hans D. Dossenbach), 1.11a (Tony Beamish), 6.7 (P. Morris), 9.3e (Bob Gibbons). Euan N. K. Clarkson: 5.28. Bruce Coleman: 10.3a (Gerald Cubitt). W. D. Hamilton: 10.1, 10.2, 10.4, 10.5, 10.7. Ole Munk: 5.31. NHPA: 6.1 (James Carmichael Jr). Chris O'Toole: 1.6a and b. Oxford Scientific Films: 1.4 (Rudie Kuiter), 2.1 (Densey Clyne), 5.19 (Michael Leach), 5.19b (J. A. L. Cooke), 10.2b (K. Jell), 10.3b (David Cayless). Portech Mobile Robotics Laboratory, Portsmouth: 9.2. Prema Photos: 8.5 (K. G. PrestonMafham). David M. Raup: 6.9. Science Photo Library: 9.3a (A. B. Dowsett), 9.3b (John Bavosi), 9.3c (Manfred Kage), 9.3d (David Patterson), 9.6 (J. C. Revy). Dr Fritz Vollrath: 2.2, 2.3, 2.4, 2.10, 2.11, 2.12, 2.13. Zefa: 9.1.
1.1 from Michell, J. (1978) Simulacra. London: Thames and Hudson. 2.5 from Hansell (1984).
2.7 and 2.8 from Robinson (1991).
2.14 and 2.15 from Terzopoulos et al. (1995) © 1995 by the Massachusetts Institute of Technology.
3.2 courtesy of the Hamilton Spectator, Canada.
4.1 courtesy of J. T. Bonner 1965, © Princeton University Press. {xi}
5.2 from Dawkins (1986) (drawing by Bridget Peace).
5.4a, b and d, 5.8a–e, 5.24a and b from Land (1980) (redrawn from Hesse, 1899).
5.4c from Salvini-Plawen and Mayr (1977) (after Hesse, 1899).
5.16a and b Hesse from Untersuchungen uber die organe der Lichtempfindung bei niederen thieren, Zeitschrift Wissenschaftliche Zoologie, 1899.
5.17, 5.19d and e, 5.25, 5.26 courtesy of M. F. Land. 5.18a and f, 5.27, 5.30 drawings by Nigel Andrews.
5.22 drawing by Kuno Kirschfeld, reproduced by permission of Naturwissenschaftliche Rundschau, Stuttgart.
5.23 courtesy of Dan E. Nilsson from Stavenga and Hardie (eds.) (1989). 5.29a–e courtesy of Walter J. Gehring et al., from Georg Haider et al. (1995).
6.16 from Meinhardt (1995).
7.2, 7.4, 7.5, 7.6, 7.7 from Ernst Haeckel (1904) Kunstformen der Natur. Leipzig and Vienna: Verlag des Bibliographischen Instituts.
7.15b from Raff and Kaufman (1983) (after Y. Tanaka, ‘Genetics of the Silkworm’, in Advances in Genetics 5: 239–317, 1953).
8.4 from Wilson (1971) (from Wheeler, 19
10, after F. Dahl).
9.4 Jean Dawkins.
9.5 © K. Eric Drexler, Chris Peterson and Gayle Pergamit. All rights reserved. Reprinted with permission from Unbounding the Future: The Nanotechnology Revolution. William Morrow, 1991.
* * *
{xii}
Climbing mount improbable
* * *
{1}
* * *
>>
CHAPTER 1
I HAVE JUST LISTENED TO A LECTURE IN WHICH THE topic for discussion was the fig. Not a botanical lecture, a literary one. We got the fig in literature, the fig as metaphor, changing perceptions of the fig, the fig as emblem of pudenda and the fig leaf as modest concealer of them, ‘fig’ as an insult, the social construction of the fig, D. H. Lawrence on how to eat a fig in society, ‘reading fig’ and, I rather think, ‘the fig as text’. The speaker's final pensée was the following. He recalled to us the Genesis story of Eve tempting Adam to eat of the fruit of the tree of knowledge. Genesis doesn't specify, he reminded us, which fruit it was. Traditionally, people take it to be an apple. The lecturer suspected that actually it was a fig, and with this piquant little shaft he ended his talk.
This kind of thing is the stock-in-trade of a certain kind of literary mind, but it provokes me to literal-mindedness. The speaker obviously knew that there never was a Garden of Eden, never a tree of knowledge of good and evil. So what was he actually trying to say? I suppose he had a vague feeling that ‘somehow’, ‘if you will’, ‘at some level’, ‘in some sense’, ‘if I may put it this way’ it is somehow ‘right’ that the fruit in the story ‘should’ have been a fig. But enough of this. It is not that we should be literalist and Gradgrindian, but our elegant lecturer was missing so much. There is genuine paradox and real poetry lurking in the fig, with subtleties to exercise an inquiring mind and {3} wonders to uplift an aesthetic one. In this book I want to move to a position where I can tell the true story of the fig. But the fig story is only one out of millions that all have the same Darwinian grammar and logic — albeit the fig story is among the most satisfyingly intricate in evolution. To anticipate the central metaphor of the book, the fig tree stands atop one of the highest peaks on the massif of Mount Improbable. But peaks as high as the fig's are best conquered at the end of the expedition. Before that there is much that needs to be said, a whole vision of life that needs to be developed and explained, puzzles that must be solved and paradoxes that must be disarmed.
As I said, the story of the fig is, at the deepest level, the same story as for every other living creature on this planet. Though they differ in surface detail, all are variations on the theme of DNA and the 30 million ways by which it propagates itself. On our route we shall have occasion to look at spider webs — at the bewildering, though unconscious, ingenuity with which they are made and how they work. We shall reconstruct the slow, gradual evolution of wings and of elephant trunks. We shall see that ‘the’ eye, legendarily difficult though its evolution sometimes seems, has actually evolved at least forty and probably sixty times independently all around the animal kingdom. We shall program computers to assist our imagination in moving easily through a gigantic museum of all the countless creatures that have ever lived and died, and their even more numerous imaginary cousins, who have never been born. We shall wander the paths of Mount Improbable, admiring its vertical precipices from afar, but always restlessly seeking the gently graded slopes on the other side. The meaning of the parable of Mount Improbable will be made clear, and much else besides. I need to begin by clarifying the problem of apparent design in nature, its relationship to true, human design and its relationship to chance. This is the purpose of Chapter 1.
The Natural History Museum in London has a quirky collection of stones that chance to resemble familiar objects: a boot, a hand, a baby's skull, a duck, a fish. They were sent in by people who genuinely suspected that the resemblance might mean something. But ordinary stones weather into such a welter of shapes, it is not surprising if occasionally we find one that calls to mind a boot, or a duck. Out of all {4} the stones that people notice as they walk about, the museum has preserved the ones that they pick up and keep as curiosities. Thousands of stones remain uncollected because they are just stones. The coincidences of resemblance in this museum collection are meaningless, though amusing. The same is true when we think we see faces, or animal shapes, in clouds or cliff profiles. The resemblances are accidents. The craggy hillside in Figure 1.1 is supposed to suggest the profile of the late President Kennedy. Once you have been told, you can just see a slight resemblance to either John or Robert Kennedy. But some don't see it and it is certainly easy to believe that the resemblance is accidental. You couldn't, on the other hand, persuade a reasonable person that Mount Rushmore, in South Dakota, had just happened to weather into the features of Presidents Washington, Jefferson, Lincoln and Theodore Roosevelt. We do not need to be told that these were deliberately carved (under the direction of Gutzon Borglum). They are obviously not accidental: they have design written all over them.
Figure 1.1 A pure accident. President Kennedys profile in a hillside in Hawaii. {5}
The difference between Mount Rushmore and the weathered likeness of John Kennedy (or Mont St Pierre in Mauritius or other such curiosities of natural weathering) is this. The sheer number of details in which the Mount Rushmore faces resemble the real things is too great to have come about by chance. The faces are clearly recognizable, moreover, when seen from different angles. Figure 1.1's chance resemblance to President Kennedy, on the other hand, is noticed only if the cliff is seen from a particular angle and in a particular light. Yes, a rock can weather into the shape of a nose seen from a certain vantage point, and maybe a couple of other rocks happen to have tumbled into the shape of lips. It is not much to ask of chance that it should produce a modest coincidence like this, especially if the photographer has every possible angle to choose from and only one gives the resemblance (and there is the added fact, which I'll return to in a moment, that the human brain seems actively eager to see faces: it seeks them out). But Mount Rushmore is another matter. Its four heads are clearly designed. A sculptor conceived them, drew them out on paper, made meticulous measurements all over the cliff, and supervised teams of workmen who wielded pneumatic drills and dynamite to carve out the four faces, each sixty feet high. The weather could have done the same job as the artfully deployed dynamite. But of all the possible ways of weathering a mountain, only a tiny minority would be speaking likenesses of four particular human beings. Even if we didn't know the history of Mount Rushmore, we'd estimate the odds against its four heads being carved by accidental weathering as astronomically high — like tossing a coin forty times and getting heads every time.
I think that the distinction between accident and design is clear, in principle if not always in practice, but this chapter will introduce a third category of objects which is harder to distinguish. I shall call them designoid (pronounced ‘design-oid’ not ‘dezziggnoid’). Designoid objects are living bodies and their products. Designoid objects look designed, so much so that some people — probably, alas, most people — think that they are designed. These people are wrong. But they are right in their conviction that designoid objects cannot be the result of chance. Designoid objects are not accidental. They have in fact been {6}
shaped by a magnificently non-random process which creates an almost perfect illusion of design.
Figure 1.2 shows a living sculpture. Beetles in general don't look like ants. So, if I see a beetle that looks almost exactly like an ant — a beetle, moreover, that makes its living entirely in an ants’ nest — I shall rightly suspect that the coincidence means something. The top animal is actually a beetle — its closer cousins are common or garden beetles — but it looks like an ant, walks like an ant, and lives among ants in an ants’ nest. The one at the bottom is a real ant. As with any realistic statue, the resemblance to the model is not an accident. It demands an explanation other than s
heer chance. What kind of an explanation? Since all beetles that look strikingly like ants live in ants’ nests, or at least in close association with ants, could it be some chemical substance from the ants, or some infection from the ants, rubbing off on the beetles and changing the way they grow? No, the true explanation — Darwinian natural selection — is very different, and we shall come to it later. For the moment, it is enough that we are sure this resemblance, and other examples of ‘mimicry’, are not {7}
Figure 1.3 (a) A real termite, Amitermes hastatus; (b) a beetle, Coatonachthode: ovambohndicus, mimicking a termite; (c) how the trick is done.
accidental. They are either designed or they are due to some process that produces results just as impressive as design. We shall look at some other examples of animal mimicry, leaving open, for the moment, the explanation of how these remarkable resemblances come about.
The previous example shows what a good job beetle flesh can do if it ‘sets out to mimic’ a different kind of insect. But now look at the creature in Figure 1.3b. It appears to be a termite. Figure 1.3a is a real termite, for comparison. The specimen in Figure 1.3b is an insect, but it is not a termite. It is, in fact, a beetle. I admit that I've seen better mimics in the insect world, including the ant-mimicking beetle of the previous example. The ‘beetle’ here is just a little odd. Its legs seem to lack proper joints, like little twisty balloons. Since, like any other insect, a beetle has jointed legs at its disposal, you might hope for a better shot at mimicking a termite's jointed legs. So, what is the solution to this conundrum? Why does this mimic look like an inflated dummy rather than like a real, jointed insect? The answer can be seen in Figure 1.3c, which is one of the most astonishing spectacles in all natural history. It shows the termite-mimicking beetle in side view. The true head of the beetle is a diminutive affair (you can see the eye just near the normal, jointed antennae), attached to a slender trunk or thorax bearing three normal, jointed beetle legs, on which it actually {8} walks. It is with the abdomen that the trick is done. It is arched backwards so that it hangs over and completely covers the head, thorax and legs like a parasol. The entire ‘termite’ is constructed from the (anatomically) rear half of the beetle's abdomen. The ‘termite head’ is the rear tip of the beetle's abdomen. The ‘termite legs’ and ‘antennae’ are flapping excrescences of the abdomen. No wonder the quality of the mimicry is not quite up to the standard of the beetle's ant-mimicking cousin of the previous picture. This termite-mimicking beetle, by the way, lives in termite nests, making its living as a parasite in much the same way as Figure 1.2's ant-mimicking beetle makes its living among ants. Although the quality of the resemblance is less, when you consider its starting materials the termite-mimicking beetle seems to achieve a more impressive feat of sculpture than the ant mimicking beetle. This is because the ant-mimic does it by modifying each bit of its body to look like the corresponding bit of the ant's body. But the termite-mimic does it by modifying a completely different bit of itself — the abdomen — to look like all the bits of the termite.