Such mutants are pretty freakish and unlikely to survive in nature, which is another way of saying that evolution is unlikely to incorporate homeotic mutations. I therefore did a double-take when I was walking (rather hurriedly, for I am squeamish about food) past a table piled high with leggy seafood at a banquet in Australia. The animal that caught my attention is known to Australian gourmets as a deepwater bug. It is a kind of lobster, member of a group known variously around the world as slipper lobsters, Spanish lobsters and shovelnosed lobsters. Figure 7.16 shows a typical specimen, borrowed from the Oxford Museum, of the genus Scyllarus. What strikes me about these animals is that they appear to have two rear ends. The illusion results from the fact that the antennae at the front (strictly the second pair of antennae) look just like the appendages called uropods which are the most prominent feature of the rear end of any lobster. I do not know why the antennae are this shape. It may be that they are
Figure 7.16 Could this represent a homeotic mutant that has been evolutionarily successful in nature? Scyllarus, a shovelnosed lobster. {252}
used as shovels, or it may be that predators are fooled by the same illusion as impressed me. Lobsters have an exceedingly fast withdrawal reflex, using a giant nerve cell dedicated to the purpose. They shoot backwards with bewildering speed when threatened. A predator might anticipate this reflex by aiming behind the lobsters present position. This might pay off with an ordinary lobster, but with Scyllarus the apparent ‘behind’ could well turn out to be the front, and the second-guessing predator would have pounced in precisely the wrong direction. Whether or not this particular speculation is justified, these lobsters presumably derive some benefit from their oddly shaped antennae. Whatever that benefit is, I shall press these animals into the service of a more foolhardy speculation. My suggestion is that Scyllarus may actually present an example in the wild of a homeotic mutation, analogous to antennapedia in Drosophila in the laboratory. Unlike antennapedia, this mutation has been incorporated into an actual evolutionary change in nature. My tentative conjecture is that an ancestral Scylland mutated homeotically, slipping the developmental sub-routine appropriate to a uropod into a segment where an antenna ought to be, and that the change conferred some benefit. If I am right, it would constitute a rare example of a macro-mutation's being favoured by natural selection: a rare vindication of the so-called ‘hopeful monster’ theory that we met in Chapter 3.
That is all very speculative. Homeotic mutations definitely occur in the laboratory, and embryologists draw upon their inspiration to build a detailed picture of the mechanics of development of the segmental body plan of arthropods. Fascinating as these details are, they are beyond the scope of this chapter. I'll conclude by inviting the reader to contemplate some real arthropods in the light of the computer arthromorphs and their three-tiered kaleidoscopic mutations.
Look at the real arthropods of Figure 7.17, and imagine how they might have evolved their form through arthromorph-style kaleidoscopic genes. Do any of them, for instance, have the tapering pattern that we saw in Figure 7.13e? Now, again looking at the real arthropods, imagine a mutation that changes some small detail of the tips of limbs, or some detail of the shape of the trunk region of segments. First think about your imagined mutation applying to only a {253}
Figure 7.17 What can be done with segments: a sampling of arthropods. Clockwise from top left: four crustaceans, a eurypterid (extinct giant ‘sea scorpion, which could easily reach a length of nearly three metres), and a palpigrade (distant relation of spiders and scorpions).
single segment. My guess is that you have automatically imagined it mirrored on the left and right side, but this is not inevitable. It is, in itself, an example of kaleidoscopic embryology. Now think of a mutation affecting the tips of limbs, but this time the limbs of a sequence of adjacent segments. The animals of Figure 7.17 show several examples of sequences of adjacent segments that resemble each other. Third, think of a similar mutation but affecting the tips of limbs in all segments of the body (all segments that have limbs at all, that is). I find that the experience of thinking about arthromorphs and their three-tiered kaleidoscopic embryology makes me see real arthropods, like the ones in Figure 7.17, as if through new eyes. Moreover, as in the case of mirrors of symmetry, it is easy to imagine that embryologies with arthromorph-style kaleidoscopic ‘restrictions’ might prove paradoxically richer in evolutionary {254} potential than more lax, unrestricted embryologies. The shapes in Figure 7.17, and those of countless other arthropods not illustrated here, seem to me to make a special kind of sense in the light of this way of thinking.
The central message of this chapter is that kaleidoscopic embryologies, whether working through segments and clusters of segments arranged in a line from front to rear as in an insect, or through ‘mirrors’ of symmetry as in a jellyfish, are paradoxically both restrictions and enhancements. They restrict evolution in that they limit the range of variation available for selection to work upon. They enhance evolution in that — to put it in language that personifies selection forgivably — they save natural selection from wasting its time exploring vast regions of search space which are never going to be any good anyway. The world is populated by major groups of animals — arthropods, molluscs, echinoderms, vertebrates — each one of which has a form of kaleidoscopically restricted embryology which has proved evolutionarily fruitful. Kaleidoscopic embryologies have what it takes to inherit the earth. Whenever a major shift in kaleidoscopic mode or ‘mirror’ has spawned a successful evolutionary radiation, that new mirror or mode will be inherited by all the lineages in that radiation. This is not ordinary Darwinian selection but it is a kind of high-level analogy of Darwinian selection. It is not too fanciful to suggest as its consequence that there has been an evolution of improved evolvability. {255}
<<
* * *
>>
CHAPTER 8
I WAS DRIVING THROUGH THE ENGLISH COUNTRYSIDE with my daughter Juliet, then aged six, and she pointed out some flowers by the wayside. I asked her what she thought wildflowers were for. She gave a rather thoughtful answer. ‘Two things,’ she said. ‘To make the world pretty, and to help the bees make honey for us.’ I was touched by this and sorry I had to tell her that it wasn't true.
My little girl's answer was not too different from the one that most adults, throughout history, would have given. It has long been widely believed that brute creation is here for our benefit. The first chapter of Genesis is explicit. Man has ‘dominion’ over all living things, and the animals and plants are there for our delight and our use. As the historian Sir Keith Thomas documents in his Man and the Natural World, this attitude pervaded medieval Christendom and it persists to this day. In the nineteenth century, the Reverend William Kirby thought that the louse was an indispensable incentive to cleanliness. Savage beasts, according to the Elizabethan bishop James Pilkington, fostered human courage and provided useful training for war. Horseflies, for an eighteenth-century writer, were created so ‘that men should exercise their wits and industry to guard themselves against them’. Lobsters were furnished with hard shells so that, before eating them, we could benefit from the improving {256} exercise of cracking their claws. Another pious medieval writer thought that weeds were there to benefit us: it is good for our spirit to have to work hard pulling them up.
Animals have been thought privileged to share in our punishment for Adams sin. Keith Thomas quotes a seventeenth-century bishop on the point: ‘Whatsoever change for the worse is come upon them is not their punishment, but a part of ours.’ This must, one feels, be a great consolation to them. Henry More, in 1653, believed that cattle and sheep had only been given life in the first place so as to keep their meat fresh ‘till we shall have need to eat them’. The logical conclusion to this seventeenth-century train of thought is that animals are actually eager to be eaten.
The pheasant, partridge and the lark
Flew to thy house, as to the Ark.
The willing ox of hims
elf came
Home to the slaughter, with the lamb;
And every beast did thither bring
Himself to be an offering.
Douglas Adams developed this conceit to a futuristically bizarre conclusion in The Restaurant at the End of the Universe, part of the brilliant Hitchhiker's Guide to the Galaxy saga. As the hero and his friends sit down in the restaurant, a large quadruped obsequiously approaches their table and in pleasant, cultivated tones offers itself as the dish of the day. It explains that its kind has been bred to want to be eaten and with the ability to say so clearly and unambiguously: ‘Something off the shoulder, perhaps? ... Braised in a white wine sauce? ... Or the rump is very good ... I've been exercising it and eating plenty of grain, so there's lots of good meat there.’ Arthur Dent, the least galactically sophisticated of the diners, is horrified but the rest of the party order large steaks all round and the gentle creature gratefully trots off to the kitchen to shoot itself (humanely, it adds, with a reassuring wink at Arthur).
Douglas Adams's story is avowed comedy but, to the best of my belief, the following discussion of the banana, quoted verbatim {257} from a modern tract kindly sent by one of my many creationist correspondents, is intended seriously.
Note that the banana:
1. Is shaped for human hand
2. Has non-slip surface
3. Has outward indicators of inward contents: Green — too early; Yellow — just right; Black — too late
4. Has a tab for removal of wrapper
5. Is perforated on wrapper
6. Biodegradable wrapper
7. Is shaped for mouth
8. Has a point at top for ease of entry
9. Is pleasing to taste buds
10. Is curved towards the face to make eating process easy.
The attitude that living things are placed here for our benefit still dominates our culture, even where its underpinnings have disappeared. We now need, for purposes of scientific understanding, to find a less human-centred view of the natural world. If wild animals and plants can be said to be put into the world for any purpose — and there is a respectable figure of speech by which they can — it surely is not for the benefit of humans. We must learn to see things through non-human eyes. In the case of the flowers with which we began our discussion, it is at least marginally more sensible to see them through the eyes of bees and other creatures that pollinate them.
The whole life of bees revolves around the colourful, scented, nectar-dripping world of flowers. I am not just talking about honey-bees, for there are thousands of different species of bee and they all depend utterly on flowers. Their larvae are fed on pollen, while the exclusive fuel for their adult flight-motors is nectar which is also entirely provided for them by flowers. When I say ‘provided for them’ I mean it in slightly more than an idle sense. Pollen, unlike nectar, is not provided purely for them, because the plants make pollen mainly for their own purposes. The bees are welcome to eat some of the pollen because they provide such a valuable service in carrying pollen from one flower to another. But nectar is a more extreme case. It {258} doesn't have any other raison d'être than to feed bees. Nectar is manufactured, in large quantities, purely for bribing bees and other pollinators. The bees work hard for their nectar reward. To make one pound of clover honey, bees have to visit about ten million blossoms.
'Flowers,’ the bees might say, ‘are there to provide us bees with pollen and nectar.’ Even the bees haven't got it quite right. But they are a lot more right than we humans are if we think that flowers are there for our benefit. We might even say that flowers, at least the bright and showy ones, are bright and showy because they have been ‘cultivated’ by bees, butterflies, hummingbirds and other pollinators. The original lecture upon which this chapter is based was called ‘The Ultraviolet Garden. This was a parable. Ultraviolet light is a kind of light that we can't see. Bees can, and they see it as a distinct colour, sometimes called bee purple. Flowers are bound to look very different through the eyes of bees (Figure 8.1). And in just the same way, the question ‘What are flowers good for?’ is a question that we are better off examining through the eyes of bees rather than through human eyes.
Figure 8.1 (a) evening primrose, Oenothra, photographed using visible (by humans) light; (b) the same, photographed by ultra-violet light (which insects can see but we can't) to show the star-shaped pattern in the centre. Presumably this pattern helps guide insects to the nectar and pollen. {259}
'The Ultraviolet Garden plays on the strangeness of bee vision only as a parable for changing our point of view about who or what it is that flowers — and all other living creatures — are ‘for the good of. If flowers had eyes, their view of the world might seem even odder to us than the alien ultraviolet visions of bees. How would bees appear through vegetable eyes? What are bees good for, from the point of view of the flowers? They are guided missiles for firing pollen from one flower to another. The background to this needs an explanation.
First, there are in general good genetic reasons for preferring cross-fertilization by pollen from a different plant. Incestuous self-mating would lose the benefits of sexual reproduction (whatever they are, which is an interesting question in itself). A tree that pollinated its female flowers with pollen from its own male flowers might almost as well not bother to pollinate at all. It would be more efficient to produce a vegetative clone of itself. Many plants of course do just this, and there is something to be said for it. But as we saw earlier there are also conditions in which there is even more to be said for reshuffling one's genes with those of another individual. It would require a massive digression to explain the detailed arguments, but there must be some substantial benefits to playing sexual roulette, otherwise natural selection wouldn't permit it to be such a driving obsession amongst almost all of animal and plant life. Whatever those benefits are, they would largely vanish if, instead of shuffling your genes with those of another individual, you simply shuffled them with a second, identical set of your own genes.
Flowers have no role in the life of their plant other than to exchange genes with another plant that has a different hand of genes. Some, like grasses, do it by wind. The air is lavishly flooded with pollen, a tiny proportion of which is lucky enough to drift on to the female parts of a flower of the same species (another proportion of it drifts into the noses and eyes of hayfever sufferers). This method of pollination is haphazard and, from some points of view, wasteful. It is often more efficient to exploit the wings and muscles of insects (or other vectors such as bats or humming-birds). This technique aims the pollen much more directly at its target, and consequently far less pollen is needed. On the other hand there has to be some expenditure {260}
Figure 8.2 Insect-mimicking orchid.
Iberian Ophrys, Ophrys vernixia.
on luring the insects. Part of the budget goes on advertising — bright-coloured petals and powerful scents. Part goes in bribes of nectar.
Nectar is high-quality aviation-fuel for an insect and it is costly for a plant to manufacture. Some plants duck out of the expense and employ deceptive advertising instead. Most famous are those orchids whose flowers look and smell like female insects. Male insects attempt to (Figure 8.2) copulate with the flowers and are inadvertently loaded with pollen bundles, or, at the other end of the trail, relieved of their pollen bundles. There are bee orchids that mimic female bees, and equally specialized fly orchids and wasp orchids. One of the wasp mimics, the wellnamed hammer orchid, keeps its dummy female wasp on the end of a hinged and spring-loaded stalk, cocked a fixed distance away from the pollen-bearing part of the flower (Figure 8.3). When the male wasp lands on the female dummy the spring is released. The male wasp is slammed, violently and repeatedly, against the anvil where the pollen sacs are kept. By the time the male wasp shakes himself free, his back is loaded with two pollen sacs.
Every bit as ingenious is the so-called bucket orchid, which works a little like a pitcher plant but with an important difference. The f
lower contains a large pool of liquid, alluringly scented to smell like the sexual attractant secreted by the females of a particular species of bee. {261}
Figure 8.3 Hammer orchid, Drakaea fitzgeraldii: (a) the wasp alights on the lure; the hinge buckles, slamming the wasp's back repeatedly against the pollinia.
A male of this species is attracted to the liquid, falls in and nearly drowns. The only escape is through a narrow tunnel. This the struggling bee eventually discovers and he crawls through it to salvation. At the far end of the tunnel there is a complicated gateway in which he is trapped for several minutes before he can wriggle free. During this final struggle at the portal of the tunnel, two large round pollen sacs are neatly transferred to his back. He then flies off and — sadder perhaps, but not wiser — falls into another bucket orchid. He again nearly drowns, again painfully pushes through the escape-tunnel and again is held for a while at the exit to freedom. During this period the second orchid relieves him of the pollen sacs and pollination is complete.
Never mind that ‘sadder but not wiser’. As ever, the temptation to impute conscious intention should be resisted. It is, if anything, more tempting for the case of the plant. On both sides, the correct way to think of what is going on is in terms of unconsciously crafted machinery. Pollen that contains genes for building bee-manipulating bucket orchids is carried by bees. Pollen that contains genes for building orchids that are less accomplished at controlling bee behaviour is less likely to be carried by bees. So, as the generations go by, orchids get better at manipulating bees (although, actually, it has to be admitted that bee orchids are not in practice spectacularly successful at actually fooling bees into copulating with them).