The Blind Watchmaker (55 page)

30. Eldredge, N. (1985)
Unfinished Synthesis: biological hierarchies and modern evolutionary thought
. New York: Oxford University Press.

31. Fisher, R.A. (1930)
The Genetical Theory of Natural Selection
. Oxford: Clarendon Press. 2nd edn paperback. New York: Dover Publications.

32. Gillespie, N.C. (1979)
Charles Darwin and the Problem of Creation
. Chicago: University of Chicago Press.

33. Goldschmidt, R.B. (1945) Mimetic polymorphism, a controversial chapter of Darwinism.
Quarterly Review of Biology
, 20: 147-64 and 205-30.

34. Gould, S. (1980)
The Panda’s Thumb
. New York: W.W.Norton.

35. Gould, S.J. (1980) Is a new and general theory of evolution emerging?
Paleobiology
, 6: 119-30.

36. Gould, S.I. (1982) The meaning of punctuated equilibrium, and its role in validating a hierarchical approach to macroevolution. In R.Milkman (ed.)
Perspectives on Evolution
, pp. 83-104. Sunderland, Mass: Sinauer.

37. Gribbin, J. & Cherfas, (1982)
The Monkey Puzzle
. London: Bodley Head.

38. Griffin, D.R. (1958)
Listening in the Dark
. New Haven: Yale University Press.

39. Hallam, A. (1973)
A Revolution m the Earth Sciences
. Oxford: Oxford University Press.

40. Hamilton, W.D. & Zuk, M. (1982) Heritable true fitness and bright birds: a role for parasites?
Science
, 218: 384-7.

41. Hitching, F. (1982)
The Neck of the Giraffe, or Where Darwin Went Wrong
. London: Pan.

42. Ho, M.W. & Saunders, P. (1984)
Beyond Neo-Darwinism
. London: Academic Press.

43. Hoyle, F. & Wickramasinghe, N.C. (1981)
Evolution from Space
. London: J.M.Dent.

44. Hull, D.L. (1973)
Darwin and his Critics
. Chicago: Chicago University Press.

45. Jacob, F. (1982)
The Possible and the Actual
. New York: Pantheon.

46. Yerison, H.J. (1985) Issues in brain evolution. In R.Dawkins &. M.Ridley (eds) Oxford Surveys in Evolutionary Biology, 1: 102-34.

47. Kimura, M. (1982)
The Neutral Theory of Molecular Evolution
. Cambridge: Cambridge University Press.

48. Kitchen P. (1983)
Abusing Science: the case against creationism
. Milton Keynes: Open University Press.

49. Land, M.F. (1980) Optics and vision in invertebrates. In H.Autrum (ed.)
Handbook of Sensory Physiology
, pp. 471-592. Berlin: Springer.

50. Lande, R. (1980) Sexual dimorphism, sexual selection, and adaptation in polygenic characters.
Evolution
, 34: 292-305.

51. Lande, R. (1981) Models of speciation by sexual selection of polygenic traits.
Proceedings of the National Academy of Sciences
, 78: 3721-5.

52. Leigh, E.G. (1977) How does selection reconcile individual advantage with the good of the group?
Proceedings of the National Academy of Sciences
, 74: 4542-6.

53. Lewontin, R.C. & Levins, R. (1976) The Problem of Lysenkoism. In H.&S.Rose (eds)
The Radicalization of Science
. London: Macmillan.

54. Mackie, L. (1982)
The Miracle of Theism
. Oxford: Clarendon Press.

55. Margulis, L. (1981)
Symbiosis in Cell Evolution
. San Francisco: W.H.Freeman.

56. Maynard Smith, J. (1983) Current controversies in evolutionary biology. In M.Grene |ed.)
Dimensions of Darwinism
, pp. 273-86. Cambridge: Cambridge University Press.

57. Maynard Smith, J. (1986)
The Problems of Biology
. Oxford: Oxford University Press.

58. Maynard Smith, J. et
al
. (1985) Developmental constraints and evolution.
Quarterly Review of Biology
, 60: 265-87.

59. Mayr, E. (1963)
Animal Species and Evolution
. Cambridge, Mass; Harvard University Press.

60. Mayr, E. (1969)
Principles of Systematic Zoology
. New York: McGraw-Hill.

61. Mayr, E. (1982)
The Growth of Biological Thought
. Cambridge, Mass: Harvard University Press.

62. Monod, J. (1972)
Chance and Necessity
. London: Fontana.

63. Montefiore, H. (1985)
The Probability of God
. London: SCM Press.

64. Morrison, P., Morrison, P., Eames, C. & Eames, R. (1982) Powers
of Ten
. New York: Scientific American.

65. Nagel, T. (1974) What is it like to be a bat?
Philosophical Review
, reprinted in D.R.Hofstadter & D.C.Dennett (eds). The Mind’s I, pp. 391-403, Brighton: Harvester Press.

66. Nelkin, D. (1976) The science textbook controversies.
Scientific American
234 (4): 33-9.

67. Nelson, G. & Platnick, N.I. (1984) Systematics and evolution, m M-W Ho & P.Saunders (eds).
Beyond Neo-Darwinism
. London: Academic Press.

68. O’Donald, P. (1983) Sexual selection by female choice. In P.P.G.Bateson (ed.)
Mate Choice
, pp. 53-66. Cambridge: Cambridge University Press.

69. Orgel, L.E. (1973)
The Origins of Life
. New York: Wiley.

70. Orgel, L.E. |197y] Selection in vitro.
Proceedings of the Royal Society of London
, B, 205: 435-^2.

71. Paley, W. (1828)
Natural Theology
, 2nd edn. Oxford: I.Vincent.

72. Penney, D., Foulds, L.R. & Hendy, M.D. (1982) Testing the theory of evolution by comparing phylogenetic trees constructed from five different protein sequences.
Nature
, 297: 197-200.

73. Ridley, M. (1982) Coadaptation and the inadequacy of natural selection.
British Journal for the History of Science
, 15: 45-68.

74. Ridley, M. (1986)
The Problems of Evolution
. Oxford: Oxford University Press.

75. Ridley, M. (1986)
Evolution and Classification: the reformation of cladism
. London: Longman.

76. Ruse, M. (1982)
Darwinism Defended
. London: Addison-Wesley.

77. Sales, G. & Pye, D. (1974)
Ultrasonic Communication by Animals
. London: Chapman &. Hall.

78. Simpson, G.G. (1980)
Splendid Isolation
. New Haven: Yale University Press.

79. Singer, P. (1976)
Animal Liberation
. London: Cape.

80. Smith, J.L.B. (1956)
Old Fourlegs: the story of the Coelacanth
. London: Longmans, Green.

81. Sneath, P.H.A. & Sokal, R.R. J1973)
Numerical Taxonomy
. San Francisco: W.H.Freeman.

82. Spiegelman, S. (1967) An
in vitro
analysis of a replicating molecule.
American Scientist
, 55: 63-8.

83. Stebbins, G.L. (1982)
Darwin to DNA, Molecules to Humanity
. San Francisco: W.H.Freeman.

84. Thompson, S.P. (1910)
Calculus Made Easy
. London: Macmillan.

85. Trivers, R.L. (1985)
Social Evolution
. Menio Park: Benjamin-Cummings.

86. Turner, R.G. (1983)
The hypothesis that explains mimetic resemblance explains evolution: the gradualist-saltationist schism
. In M.Grene (ed.)
Dimensions of Darwinism
, pp. 129^9. Cambridge: Cambridge University Press.

87. Van Valen, L. (1973| A new evolutionary law.
Evolutionary Theory
, 1: 1-30.

88. Watson, D. (1976)
Molecular Biology of the Gene
. Menio Park: Benjamin-Cummings.

89. Williams, G.C. (1966)
Adaptation and Natural Selection
. New Jersey: Princeton University Press.

90. Wilson, E.O. (1971)
The Insect Societies
. Cambridge, Mass: Harvard University Press.

91. Wilson, E.O. (1984)
Biophilia
. Cambridge, Mass: Harvard University Press

92. Young, Z. (1950)
The Life of Vertebrates
. Oxford: Clarendon Press.

APPENDIX (1991)

COMPUTER PROGRAMS AND THE EVOLUTION OF EVOLVABILITY

The biomorph computer program described in Chapter 3 is now available for Apple Macintosh, RM Nimbus and IBM-compatible computers (see advertisement on p. 341). All three programs have the basic nine ‘genes’ necessary to produce the biomorphs illustrated in Chapter 3 and trillions more like them - or not so like them. The Macintosh version of the program also possesses a range of additional genes, producing ‘segmented’ biomorphs (with segmentation ‘gradients’) and biomorphic images reflected in various planes of symmetry. These enhancements of the biomorphic chromosome, together with a new colour version of the program now being developed for the Macintosh II and not yet released, led me to reflect on ‘the evolution of evolvability’. This new reprinting of
The Blind Watchmaker
provides an opportunity to share some of these thoughts.

Natural selection can act only on the range of variation thrown up by mutation. Mutation is described as ‘random’ but this means only that it is not systematically directed towards improvement. It is a highly nonrandom subset of all the variation that we can conceive of. Mutation has to act by altering the processes of existing embryology. You can’t make an elephant by mutation if the existing embryology is octopus embryology. That is obvious enough. What was less obvious to me until I started playing with the expanded Blind Watchmaker program is that not all embryologies are equally ‘fertile’ when it comes to fostering future evolution.

Imagine that a wide-open space of evolutionary opportunity has suddenly opened up - say a deserted continent has suddenly become available through natural catastrophe. What kinds of animals will fill the evolutionary vacuum? They will surely have to be descendants of individuals good at surviving in the post-catastrophe conditions. But more interestingly, some kinds of embryology might be especially good not just at surviving but at
evolving
. Perhaps the reason the mammals took over after the extinction of the dinosaurs is not just that mammals were good at individual
survival
in the post-dionsaur world. It may be that the mammalian way of growing a new body is also ‘good’ at throwing up a great variety of types - carnivores, herbivores, anteaters, tree-climbers, burrowers, swimmers and so on and the mammals can therefore be said to be good at
evolving
.

What has this to do with computer biomorphs? Shortly after developing the Blind Watchmaker program, I experimented with other computer programs that were the same except that they employed a different basic embryology - a different fundamental body-drawing rule upon which mutation and selection could act. These other programs, although superficially similar to Blind Watchmaker, turned out to be sadly impoverished in the range of evolutionary possibility that they offered. Evolution continually became stuck up sterile blind alleys. Degeneration seemed to be the commonest outcome of even the most carefully guided evolution. In contrast, the branching-tree embryology at the heart of the Blind Watchmaker program seemed ever-pregnant with renewable evolutionary resources; there was no tendency towards automatic degeneration as evolution proceeded - the richness, versatility, even beauty, seemed to be indefinitely refreshed as the generations flashed by.

Nevertheless, prolific and varied as the biomorphic fauna produced by the original Blind Watchmaker program was, I continually found myself coming up against apparent barriers to further evolution. If Blind Watchmaker’s embryology was so evolutionarily superior to those alternative programs, might there not be modifications, extensions to the embryological drawing rule that could make Blind Watchmaker itself even more luxuriant with evolutionary diversity? Or another way of putting the same question - could the basic chromosome of nine genes be expanded in fruitful directions?

In designing the original Blind Watchmaker program, I deliberately tried to avoid deploying my biological knowledge. My purpose was to exhibit the power of nonrandom selection of random variation. I wanted the biology, the design, the beauty, to
emerge
as a result of selection. I didn’t want to be able to accuse myself, later, of having built it in when I wrote the program in the first place. The branching-tree embryology of Blind Watchmaker was the very first embryology that I tried. That I had in fact been lucky was suggested by my subsequent disappointing experience with alternative embryologies. At all events, in thinking of ways to expand the basic ‘chromosome’, I did allow myself the luxury of using some of my biological knowledge and intuition. Among the most evolutionarily successful animal groups are those that have a
segmented
body plan. And among the most fundamental features of animal body plans are their plans of
symmetry
. Accordingly, the new genes that I added to the biomorphic chromosome controlled variations in segmentation and symmetry.

We, and all vertebrates, are segmented. This is clear in our ribs and our vertebral columns, whose repetitive nature is seen not just in the bones themselves but in the associated muscles, nerves and blood-vessels. Even our heads are fundamentally segmented, but in the adult head the segmental structure has become obscured to all but professionals schooled in embryonic anatomy. Fish are more obviously segmented than we are (think of the battery of muscles lying along the backbone of a kippered herring). In crustaceans, insects, centipedes and millipedes the segmentation is even manifest on the outside. The difference, in this respect, between a centipede and a lobster is one of homogeneity. The centipede is like a long goods-train with all the trucks almost identical to one another. The lobster is like a train with a motley variety of carriages and trucks, all basically the same and with the same jointed appendages sticking out of each. But in some cases the trucks are welded together in groups and the appendages have become large legs or pincers. In the tail region the trucks are smaller and more uniform, and their clawed side-apparatuses have become small, feathery ‘swimmerets’.

To make biomorphs segmented I did the obvious thing: I invented a new gene controlling ‘Number of Segments’, and another new gene controlling ‘Distance between Segments’. One complete old-style biomorph became a single segment of a new-style biomorph.